Ecological and Genetic Factors in the Distribution and Abundance of Larval Lake Whitefish (Coregonus clupeaformis) at Douglas Point, Lake Huron

by

Lauren M. Overdyk

A Thesis presented to The University of Guelph

In partial fulfilment of requirements for the degree of Doctor in Philosophy in Integrative Biology

Guelph, Ontario, Canada

© Lauren Overdyk, July, 2015

ABSTRACT

ECOLOGICAL AND GENETIC FACTORS IN THE DISTRIBUTION AND ABUNDANCE OF LARVAL LAKE WHITEFISH (COREGONUS CLUPEAFORMIS) AT DOUGLAS POINT, LAKE HURON

Lauren M. Overdyk Advisors: University of Guelph, 2015 Professor Stephen S. Crawford Professor Robert H. Hanner

Lake Whitefish are an ecologically, economically and culturally important fish species in the Laurentian Great Lakes. Although much research has been conducted on spawning-phase adult Lake Whitefish, little research has paid attention to the ecology of larval Lake Whitefish, especially in the source waters of Bruce Nuclear Generating Station. This PhD thesis incorporates key ecological and methodological uncertainties into understanding the effects of environmental conditions on the ecology of larval Lake Whitefish at Douglas Point, Lake Huron. The result is a set of novel ideas and the development of novel methods to help answer this question. Chapter 2 investigates the effects of environmental conditions on the distribution and abundance of zooplankton as a necessary first step in understanding the ecology of larval Lake Whitefish. Chapter 3 evaluates the consistency between DNA barcoding and visual identification methods using a case study of larval fish caught in plankton tows at Stokes Bay, Lake Huron. This evaluation strongly supports the use of DNA barcoding in combination with visual identification to improve the accuracy and precision of species identification. Chapter 4 explores genetic haplotype variation of Lake Whitefish from Lake Huron using DNA barcodes from spawning-phase Lake Whitefish collected at 28 sites around Lake Huron during Fall 2012. While this study did not detect any cryptic lineages of Lake Whitefish in Lake Huron, it did reveal the presence of rare barcode haplotypes that seem to be unique to specific sampling sites. Chapter 5 develops a novel, real-time PCR assay to specifically identify Lake Whitefish in larval fish assemblages. This technique can further increase the speed of identification of Lake Whitefish. Finally, Chapter 6 investigates the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron. Ultimately, the new knowledge of larval Lake Whitefish ecology generated in this thesis should be seriously considered by Canada/Ontario, First Nations and Industry as they work together to evaluate effects of the existing Bruce Nuclear Generating Station, and the Deep Geologic Repository for Nuclear Waste that has been proposed for construction at Douglas Point, Lake Huron.

ACKNOWLEDGEMENTS

There are many people to whom I owe a debt of gratitude over the last four years, some of who were instrumental in the success of this research and some of who were entrusted with keeping my sanity and spirits high. Firstly, I would like to thank Dr. Steve Crawford and Dr. Bob Hanner - my two academic ‘dads’ - who shared their very different areas of expertise with me, took me under their wing and taught me invaluable skills. Although the last four years have not always been easy, I am grateful for the experiences and advice you both have provided.

Secondly, I would like to thank the Saugeen Ojibway Nation (Chippewas of Nawash Unceded First Nation, Saugeen First Nation) in collaboration with Bruce Power Limited who supported this research. Specifically, I would like to thank Ryan Lauzon, (Nawash Fisheries Assessment Program) for his assistance in the field. I would also like to thank the members of my advisory committee, Dr. Neil Rooney and Dr. Dan Gillis, for being available to discuss research and life, and for always offering support. With this I would also like to thank my Crawford lab mates Kathleen Ryan, Colleen Parker, Natalie Schott, Dr. Andrew Binns and Dr. Shoshanah Jacobs for being academic sounding boards. I would also specifically like to thank Laura Trout for her support and help in preparation for my qualifying exam.

The two field seasons in this thesis (Chapters 2 and 6) would not have been possible with out the assistance of many people. I would like to thank Ashley Wincikaby and Lindsey Boyd for their hard work and dedication to the collection of my field data. Lindsey, thank you for being my purse and nerding out over fish as much as me. Your friendship and support of my work has been invaluable. I would like to thank the expertise of René Lauzon for his work on our boat and Bill McKeag from the Kincardine Marina. I would also like to thank Steve Wilson from the physics department for creating custom tow frames for both the 2013 and 2014 field seasons. I also owe a special thanks to Bill Thorne, Shannon Snyder and the staff of the Inverhuron Provincial Park. Without your assistance, kindness and hospitality, my field seasons may not have been possible. A special thanks to the members of the Lake Huron Fishing Club, especially Mike Hahn, Eugene Lo, Brian Garnet and Norm Dobson. I would like to thank Dr. Beatrix Beisner and Katherine Velghe (Université du Quebéc à Montréal – UQAM) for processing water quality samples and offering advice on field sampling design. A special thanks to Dr. Josef Ackerman and Dr. Karl Cottenie (University of Guelph) for providing advice for the completion of these two chapters.

Chapters 3, 4 and 5 would not have been possible without the assistance and guidance of the following people. For Chapter 3 specifically I would like to thank members of the Hanner Lab, including Natasha Serrao, Danielle Ondrejicka, Jeff Strohm, Amanda Naaum, Andrew Frewin and Heather Braid. Special thanks to Colette Ward and Erling Holm for sharing their expertise with Great Lakes ichthyoplankon, to Justin Angevaare for advice with statistical analyses, to Colleen Parker for assistance with fish tissue preparation, to Joan Hewer and anonymous reviewers for edits. Special thanks for AAC Genomics for sequencing and the Royal Ontario Museum. For Chapters 4 and 5

iii specifically I would like to thank Nikole Freeman, Kelly Mulligan, Grace Burke, Rebecca Eberts (University of Regina), Chris Somers (University of Regina), Wendy Lee Stott (USGS), Tim Drew (White Lake Fish Culture Station, Sharbot Lake), Cadence Cumpseth and Chris Ho (BOLD for bioinformatics support). For Chapter 5 I would also like to thank Dr. Cameron Turner and an anonymous reviewer for comments leading to a substantial improvement of this manuscript.

I would like to thank my family (Adrian, Karen and Lisen) and close friends for supporting me through my academic career, especially these last four years. The highs have been high and the lows have been low and I thank you for sticking by my side through it all. I appreciate you trying to understand what I have been doing and being excited about fish even when you weren’t. Last but not least, I thank my partner in crime, James Langdon for your unwavering support and patience in this final year. Not only did you brave the frigid waters of Lake Huron in late September for me and my research, but you were a constant in ensuring I was always at my best. Your contagious cheerleading has helped me cross the finish line, and for this I am more grateful than you will ever know.

iv TABLE OF CONTENTS

Abstract……………………………………………………………………………………ii

Acknowledgements………………………………………………………………………iii

Table of Contents………………………………...………………………………………..v

List of Figures…………………………………………………………………………...viii

List of Tables……………………………………………………………………….……xii

Chapter 1. Prologue………………………………………..…………………………….1

1.2. References………………………………………..…………………………...9

Chapter 2. Effects of environmental conditions on the distribution and abundance of zooplankton at Douglas Point, Lake Huron………………………………………..18

2.1. Abstract……………………………………………………………………...18 2.2. Introduction………………………………………………………………….19 2.3. Materials and Methods………………………………………………………24 Sample Design…………………………………………………………...24 Environmental Conditions……………………………………………….25 Water Samples…………………………………………………………...26 Laboratory Analyses……………………………………………………..26 Statistical Analyses………………………………………………………28 2.4. Results……………………………………………………………………….29 Effect of Space-Time on Environmental…..…………………………….29 Effect of Space-Time/Environment on Particle Size Frequency………...30 Effect of Space-Time/Environment on Zooplankton Size Frequency…...31 2.5. Discussion…………………………………………………………………...33 Effect of Space-Time on Environment…………………………………..33 Effect of Space-Time/Environment on Particle Size Frequency………...34 Effect of Space-Time/Environment on Zooplankton Size Frequency…...35 Hypotheses and Predictions……………………………………………...36 2.6. References…………………………………………………………………...42 2.7. Appendices Appendix 2.1. Larval Lake Whitefish (Coregonus clupeaformis) Gape Size……………………………………………………………………….65 Appendix 2.2. Larval Burbot (Lota lota)……………...…………………67 Appendix 2.3. Environmental Variables and Particle Size/Zooplankton Frequencies Observed for all Stations and Samples over weeks 1-3 for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013………..68 Appendix 2.4. Historic Sampling of phyto and nonicththyo-zooplankton at Douglas Point, Lake Huron………………………………………………73

v Appendix 2.5. Larval Lake Whitefish (Coregonus clupeaformis) Diet.…82

Chapter 3. Increased taxonomic resolutions of Laurentian Great Lakes ichthyoplankton through DNA barcoding: A case study comparison against visual identification of Stokes Bay, Lake Huron ichthyoplankton………………………….84

3.1. Abstract……………………………………………………………………...84 3.2. Introduction………………………………………………………………….85 3.3. Materials and Methods………………………………………………………88 3.4. Results……………………………………………………………………….90 3.5. Discussion…………………………………………………………………...92 Concluding Remarks……………………………………………………..96 3.6. References…………………………………………………………………...98 3.7. Appendices…………………………………………………………………109 Appendix 3.1. ROM Catalogue Numbers………………………………109

Chapter 4. Extending DNA barcoding coverage for Lake Whitefish (Coregonus clupeaformis) across the three major basins of Lake Huron……………………….111

4.1. Abstract…………………………………………………………………….111 4.2. Introduction………………………………………………………………...112 4.3. Materials and Methods……………………………………………………..114 Specimen Collection……………………………………………………114 DNA Barcoding………………………………………………………...115 Haplotype Analysis……………………………………………………..116 4.4. Results……………………………………………………………………...117 4.5. Discussion………………………………………………………………….118 4.6. References………………………………………………………………….121

Chapter 5. Real-Time PCR Identification of Lake Whitefish (Coregonus clupeaformis) in the Laurentian Great Lakes……………………………………….127

5.1. Abstract…………………………………………………………………….127 5.2. Introduction………………………………………………………………...128 5.3. Materials and Methods……………………………………………………..131 Sample Collection and DNA Extraction………………………………..131 DNA Barcoding………………………………………………………...132 Real-time PCR………………………………………………………….133 5.4. Results……………………………………………………………………...137 5.5. Discussion………………………………………………………………….139 5.6. References………………………………………………………………….142

Chapter 6. The effect of environmental conditions on the distribution and abundance of larval Lake Whitefish (Coregonus clupeaformis) in nearshore embayments at Douglas Point, Lake Huron…………………………………………156

6.1. Abstract…………………………………………………………………….156

vi 6.2. Introduction………………………………………………………………...157 6.3. Materials and Methods……………………………………………………..161 Study Design……………………………………………………………161 Species Identification…………………………………………………...165 Statistical Analyses……………………………………………………..167 6.4. Results……………………………………………………………………...168 Species Identification…………………………………………………...168 Larval Distribution and Abundance…………………………………….169 Effect of Space-Time on Environment…………………………………170 Effect of Space-Time/Environment on Ichthyoplankton Standard Density………………………………………………………….171 Serendipitous Samples of Ichthyoplankton...... 173 6.5. Discussion………………………………………………………………….173 Species Identification…………………………………………………...173 Larval Distribution and Abundance…………………………………….174 Effect of Space-Time on Environment…………………………………177 Effect of Space-Time/Environment on Ichthyoplankton Standard Density………………………………………………………….177 Serendipitous Samples of Ichthyoplankton. …………………………...179 Hypotheses and Predictions…………………………………………….181 6.6. References………………………………………………………………….186 6.7 Appendices………………………………………………………………….209 Appendix 6.1. Non-Lake Whitefish ichthyoplankton distribution and abundance…………………………………………………..…..209

Chapter 7: Epilogue…………………………………………………………………...210

7.2. References……………………………………………………………….…218

vii LIST OF FIGURES

CHAPTER 1: Prologue

Figure 1.1. Location of Douglas Point within Lake Huron (black square and inlay). Map shows location of Bruce Nuclear Generating Station sites A and B along with intakes (stars) and discharges (arrows) for each station…………………………14

CHAPTER 2: Effects of environmental conditions on the distribution and abundance of zooplankton at Douglas Point, Lake Huron.

Figure 2.1. Map of Douglas Point, Lake Huron showing the 2013 sampling transect (DP= Douglas Point, BO=Outflow for BNGS ‘B’, BI=intake for BNGS ‘B’, IH=Inverhuron Bay, MP=McRae Point). Circles represent sampling stations at 3m, 10m, 20m and 40m for each transect, with the exception of the discharge (BO) where strong currents prevented effective sampling at the 3m depth. Stars indicate location of cooling water intake for BNGS ‘A’ and ‘B’ facilities. Grey arrows indicate discharge outflows of cooling water systems…………………...51

Figure 2.2. Biplot of the Redundancy Analysis after a forward selection of the relationship between space-time variables on environmental conditions sampled at Douglas Point, Lake Huron in 2013. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2….……………………52

Figure 2.3.A-C. Water temperatures (°C) sampled at the 20m station depth for all sample depths (1,5,15m) over five weeks for transects (A) Inverhuron (IH), (B) Intake B (BI) and (C) Outflow B (BO) respectively at Douglas Point, Lake Huron sampled in 2013……………………………………………………………………………53

Figure 2.4. Uniplot of the Redundancy Analysis after a forward selection of the relationship between space-time and environmental conditions on particle size sampled at Douglas Point, Lake Huron in 2013. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2...……..54

Figure 2.5. Uniplot of the Redundancy Analysis after a forward selection of the relationship between space-time variables and environmental conditions on zooplankton frequency by size sampled at Douglas Point, Lake Huron in 2013. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2………………………………………………………………...55

Figure 2.6.A-C. Zooplankton frequency sampled at the 20m station depth for all sample depths (1,5,15m) over five weeks for transects (A) Inverhuron (IH), (B) Intake B (BI) and (C) Outflow B (BO) respectively at Douglas Point, Lake Huron sampled in 2013……………………………………………………………………………56

viii

CHAPTER 2: Appendices

Figure A2.1. Ventral view of larval Lake Whitefish (Coregonus clupeaformis) showing lower maxillary (limiting gape width) with line indicating widest part of gape. Scale is 1mm. Widths measured using Image J software………………………..66

Figure A2.2. Images of larval Burbot (Lota lota) collected on 05 July 2013 (Week 3) on the Douglas Point (DP) transect, station depth of 40m, at a sample depth of 15m in Lake Huron. Scale =1mm. (A – dorsal, B-ventral). Larval fish was identified as Lota lota by DNA Barcoding…………………………………………………67

Figure A2.3.1 A-D: Temperature (oC) observed for all stations (3, 10, 20, 40m, A-D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3……………….69

Figure A2.3.2 A-D: Dissolved oxygen (mg L-1) observed for all stations (3, 10, 20, 40m, A-D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3……………….70

Figure A2.3.3 A-D: Total particle frequency observed for all stations (3, 10, 20, 40m, A- D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3……………….71

Figure A2.3.4 A-D: Total zooplankton frequency observed for all stations (3, 10, 20, 40m, A-D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3………………………………………………………………………………..72

CHAPTER 3: Increased taxonomic resolutions of Laurentian Great Lakes ichthyoplankton through DNA barcoding: A case study comparison against visual identification of Stokes Bay, Lake Huron ichthyoplankton.

Figure 3.1 Diagram of a larval Lake Whitefish (Coregonus clupeaformis) indicating morphological characteristics, especially mensural and meristic features used in a dichotomous key for ichthyoplankton of the Great Lakes basin. Adapted from Faber 2006-13 www.fishbabies.ca……………………………………………...102

CHAPTER 4: Extending DNA barcoding coverage for Lake Whitefish (Coregonus clupeaformis) across the three major basins of Lake Huron.

Figure 4.1. DNA barcode haplotype variation found in Lake Whitefish (Coregonus clupeaformis) across North American ecoregions. Each colour represents a different haplotype. Haplotypes were included from Yukon River, Swan Lake, St. Laurence River, Sharbot Lake, and Lake Huron. The size of each pie chart is proportional to the sample size from each location; however, the most dominant haplotype (A) has been excluded from Lake Huron for the purpose of

ix visualization of the less prevalent haplotypes, but all haplotypes for Lake Huron can be seen in Figure 4.2………………………………………………………..124

Figure 4.2. DNA barcode haplotype variation found in Lake Whitefish (Coregonus clupeaformis); site numbers correspond to Table 4.1 and L1 and L2 are two sites sampled in BOLD project Stokes Bay, Ontario, Lake Whitefish [SBOLW]. (A) Haplotype network analysis of DNA barcode sequences from across North America; the size of the nodes corresponds to the number of individuals that share each haplotype; the colour of each unique haplotype corresponds to the pie charts in B; (B) Geographic distribution of haplotypes from Lake Huron at each site sampled during 2012, the size of each pie chart is proportional to the sample size from each location. MB=Main Basin, GB=Georgian Bay, NC= North Channel…………………………………………………………………………125

CHAPTER 5: Real-Time PCR identification of Lake Whitefish (Coregonus clupeaformis) in the Laurentian Great Lakes.

Figure 5.1. Standard curve generated from 10-fold serial dilutions of Lake Whitefish (Coregonus clupeaformis ) DNA from 7.1 ng/µl to 0.71 pg/µl. FAM – fluorescent reporter 6-carboxyfluorescein…………………………………………………..147

Figure 5.2. Melt curve analysis peaks for: (A) no template control; (B) non-target species Longnose Sucker (Catostomus catostomus); (C) non-target species White Sucker (Catostomus commersonii); and (D) target species Lake Whitefish (Coregonus clupeaformis). Only one distinct melt peak is present in panel D for Lake Whitefish. No secondary peaks indicate no primer dimer formation………148

Figure 5.3. Results of agarose gel electrophoresis of: (A) traditional PCR using universal fish primers for the DNA barcode region; and (B) amplification with the Coregonus- specific primers designed in this study. (A) traditional PCR for identification of Longnose Sucker (LS; Catostomus catostomus) and Lake Whitefish (LW; Coregonus clupeaformis) using universal fish primers. NTC =no template control. Bands are present for both Longnose Sucker (lane 5) and Lake Whitefish (lane 6) at the expected size of 650 bp. No bands are present for no-template controls. (B) Results of agarose gel electrophoresis of Real time PCR product using primers designed in this study. NTC=no template control, non-target in lanes 3-6 include 2 replicates of Longnose Sucker (lanes 3,4) and White Sucker (lanes 5, 6). Bands observed for Lake Whitefish (LW) in lanes 7- 10 at approximately 120 bps, which is expected length of product amplicon from Real time PCR reaction. No bands present for the no template control and non- target species………………..…………………………………………………149

CHAPTER 6: The effect of environmental conditions on the distribution and abundance of larval Lake Whitefish (Coregonus clupeaformis) in nearshore embayments at Douglas Point, Lake Huron.

x Figure 6.1.A-B. Location of transects (lines), pump stations (stars) of (A) Inverhuron Bay and Holmes Bay and (B) Baie du Doré at Douglas Point, Lake Huron in 2014. Classification of transects correspond to Table 6.1…………………………...194

Figure 6.2. Biplot of the Redundancy Analysis after a forward selection of the relationship between space-time variables on environmental conditions sampled at Douglas Point, Lake Huron in 2014. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3. ……..…………….195

Figure 6.3A-C. Time series of temperature (°C) from data loggers at 2m depth on nearshore transects in (A) Inverhuron Bays and (B,C) Baie du Doré, Douglas Point, Lake Huron, for the period May-September 2014……………………..196

Figure 6.4. Uniplot of the Redundancy Analysis after a forward selection of the relationship between space-time/environment on ichthyoplankton standard densities sampled at Douglas Point, Lake Huron in 2014. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3.…………………………………………………………………………….197

Figure 6.5. Standard densities (SD) of ichthyoplankton (# larvae 1000m-3) sampled in 2014 at Douglas Point, Lake Huron, associated with prevailing (modal) wind direction on the day of sampling. (A) Lake Whitefish (Coregonus clupeaformis), (B) all species. Open circles represent observations with SD<10, while bold numbers represent week number……………………………………………...198

Figure 6.6. Weekly wind roses for direction data collected at Gunn Point, Inverhuron, Lake Huron for the six-week period from 5 May - 21 June 2014. Bold pies in sectors represent percentage of wind station observations associated with wind moving toward the direction indicated for the sector. Note that all weeks were 7 days except for week 1 which was 11 days…………………………………...199

xi LIST OF TABLES

CHAPTER 1: PROLOGUE

Table 1.1. Historical field sampling effort and associated distribution/abundance of larval Lake Whitefish (Coregonus clupeaformis) at Douglas Point, Lake Huron from 1979-2007 excluding forebay entrainment sampling at the Bruce Nuclear Generating Station. Bay= within (In) embayment or outside (Out) of embayment. Offshore = location to shore as either near or far, where nearshore is <3m and offshore is >4m. Orient = orientation of transects as either parallel or perpendicular to shore. Depth = depth contours transects follows (m). Gear= gear used to sample ichyoplankton with associated effort as either number of tows or hours for larval traps. All=all species of ichthyoplankton sampled. LW=larval Lake Whitefish sampled. LW-SD= standard density of larval Lake Whitefish (fish/1000m3) sampled. Single data entries in a cell apply to all Region/Bay combinations. Question marks represent missing data. NA= not applicable to sampling regime……………………………………………...……15

CHAPTER 2: Effects of environmental conditions on the distribution and abundance of zooplankton at Douglas Point, Lake Huron.

Table 2.1. Dates of plankton samples for each of the five transects at Douglas Point, Lake Huron during the 2013 field season. MP= McRae Point, IH=Inverhuron Bay, BI=Bruce Intake B, BO=Bruce Outflow B, DP=Douglas Point………………..57

Table 2.2. Space-time variables and environmental conditions sampled at Douglas Point, Lake Huron in 2013 with associated codes referenced in Redundancy Analyses.58

Table 2.3. Axis summary statistics from the Redundancy Analysis of space-time variables on environmental conditions sampled at Douglas Point, Lake Huron in 2013. Adjusted explained variation if 49.4%. Correlations of environmental conditions and ordination axes. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2………………………………….59

Table 2.4. Variation partitioning analysis of the significant space-time factors on environmental conditions in Redundancy Analysis forward selection sampled at Douglas Point, Lake Huron in 2013. Percent (%) Explains is the percent (%) variation explained solely by that individual factor. p values are applied to factor as a whole. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2…………………………………………………60

Table 2.5. Axis summary statistics from the Redundancy Analysis of space-time variables and environmental conditions on particle size frequency sampled at Douglas Point, Lake Huron in 2013. Adjusted explained variation is 45.6% (R2=63.9%). Correlations of environmental conditions and ordination axes. NS=

xii not selected by analysis. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2……………………………...…...61

Table 2.6. Variation partitioning analysis of the significant space-time variables and environmental conditions on particle size frequency in Redundancy Analysis forward selection sampled at Douglas Point, Lake Huron in 2013. Percent (%) Explains is the percent (%) variation explained solely by that individual factor. p values are applied to factor as a whole. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2……………………….62

Table 2.7. Axis summary statistics from the Redundancy Analysis of space-time variables and environmental conditions on zooplankton size frequency sampled at Douglas Point, Lake Huron in 2013. Adjusted explained variation is 43.2% (R2=91.3%). Correlations of environmental conditions and ordination axes. NS= not selected by analysis. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2………………………...………...63

Table 2.8. Variation partitioning analysis of the significant space-time variables and environmental conditions on zooplankton size frequency in Redundancy Analysis forward selection sampled at Douglas Point, Lake Huron in 2013. Percent (%) Explains is the percent (%) variation explained solely by that individual factor. p values are applied to factor as a whole. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2……………………….64

CHAPTER 2: APPENDICES

Table A2.1. Gape size of larval Lake Whitefish (Coregonus clupeaformis) over eight weeks of development measured by horizontal measurement of the widest part of lower maxillary (limiting gape width) (See Figure A2.2). Where N= sample size, SD=standard deviation. Lower maxillary widths were measured using Image J software…………………………………………………………………………..65

Table A2.4. Plankton (phyto, zoo) species presence at Douglas Point, Lake Huron compiled from Johnson 1973; Wismer et al. 1986; Golder Associates 2008e. Sizes (µm), lengths and widths when available and morphological features for each genus/species provided when available. Species presenting extreme sexual dimorphism have been treated as two different species, one for male and one for female. * indicates most prominent genera present at the site…………………...73

Table A2.5. Larval Lake Whitefish (Coregonus clupeaformis) diet by class/order and genus/species in field and laboratory with associated references. *mesocosm experiment **remained quantitatively the most important prey item…………...82

xiii CHAPTER 3: Increased taxonomic resolutions of Laurentian Great Lakes ichthyoplankton through DNA barcoding: A case study comparison against visual identification of Stokes Bay, Lake Huron ichthyoplankton.

Table 3.1. Consistency of species level visual identification compared to DNA barcoding of ichthyoplankton. N= novice, E= expert. UN=unknown, FM=family, na=not attempted. Dark grey squares are incorrect identifications, light grey squares are not attempted, medium grey squares are identified correctly to family. Letters in boxes are what the identifier originally identified the specimen as…………….103

Table 3.2. Frequency and use of different morphological characteristics in Auer's (1982) dichotomous key for ichthyoplankton of the Great Lakes basin, used by novice/expert identifiers in this study to determine taxonomic identification of ichthyoplankton collected from Stokes Bay, Ontario, Lake Huron (species identification for larvae of Subfamily Coregoninae). Grey shading indicates Family/Subfamily/species in the sample, as determined by DNA barcoding…..106

CHAPTER 3: APPENDICES

Table A3.1. Royal Ontario Museum Catalogue Numbers of Voucher Specimens…….109

CHAPTER 4: Extending DNA barcoding coverage for Lake Whitefish (Coregonus clupeaformis) across the three major basins of Lake Huron.

Table 4.1. Collection site data for the Lake Whitefish (Coregonus clupeaformis) for the 2012 sampling effort in Lake Huron. Map number corresponds to the site number indicated on Figure 1; Basin name abbreviations are NC = North Channel, GB=Georgian Bay, and MB= Main Basin; the sex of fish sampled is indicated by M (male) and F (female); early designated the start of the fall spawning season, late designates the end of the fall spawning season…………………………….126

CHAPTER 5: Real-Time PCR Identification of Lake Whitefish (Coregonus clupeaformis) in the Laurentian Great Lakes.

Table 5.1. Sequences of species-specific primer and probe set for Coregonus clupeaformis. The amplicon length for the primers is 122bp of the DNA barcode region. All primers and probes target the cytochrome C oxidase subunit I (COI) DNA barcode region. FAM – fluorescent reporter 6-carboxyfluorescein. BHQ – black hole quencher 1……………………………………..……………………150

Table 5.2. Species data for sequences used in the design of the primer/probe set. Barcode of Life Data System (Bold; htt://boldsystems.org) project code indicates the projects that contained sequences used with the number of sequences taken from each project in brackets; the number of sequences for each species and haplotypes per species are also indicated. Full specimen details for sequences are available from the BOLD Dataset ‘Dataset of Coregonids’, project code DS-CORG…...151

xiv Table 5.3. Location of forward and reverse primer on the 650 base pair amplicon of the DNA barcode region of cytochrome c oxidae subunit I (COI) and corresponding probe location and alignment among Coregonus species. Created by using the NCBI-Primer BLAST making tool. **primers do not bind to the genus Prosopium, there is one base pair difference present…………………………...152

Table 5.4. DNA mixtures containing different proportions of Coregonus clupeaformis (CC) [7.1 ng/ µl] mixed with non-target species Coregonus hoyi (CH) [7.1 ng/ µl] and Sander vitreus (SV) [7.1 ng/ µl]. Absolute concentration of CC DNA in each mixture provided. Ct values for each replicate shown along with average Ct value………………………………………………………………………….…154

Table 5.5. Validation of protocol and primer and probe set on StepOnePlus Real-Time PCR machine and comparison of Ct values. All samples extracted from larval caudal fin clips, except for Salmo salar and Prosopium cylindraceum, which were collected from adult muscle samples. [DNA] in ng/ µl. CC= Coregonus clupeaformis, CH= Coregonus hoyi, SV= Sander vitreus. See Table II for [CC] DNA (ng/ µl) in mixtures 1-5. DNA concentrations of samples used listed in methods for each species…………………………………………………….....155

CHAPTER 6: The effect of environmental conditions on the distribution and abundance of larval Lake Whitefish (Coregonus clupeaformis) in nearshore embayments at Douglas Point, Lake Huron.

Table 6.1. Classification of transects by region, bay and offshore designations. Transects can be classified as in embayment or out of embayment (Bay); near or offshore (Offshore). Associated variable types correspond to space-time variables used in Redundancy Analyses. Transect codes correspond to Figure 6.1A-B………..200

Table 6.2. Dates of plankton samples for each of the three nearshore embayments at Douglas Point, Lake Huron during the 2014 field season…………………….201

Table 6.3. Space-time variables, environmental conditions and response variables sampled at Douglas Point, Lake Huron in 2014 with associated codes referenced in Redundancy Analyses tables and figures…………………………………..202

Table 6.4. Consistency of Lake Whitefish (Coregonus clupeaformis) specific Real-time PCR (qPCR) compared to DNA barcode species identification of all ichthyoplankton caught at Douglas Point, Lake Huron in 2014. Real-time PCR (qPCR) threshold cycle (Ct) value indicates a relative measure of target species in a sample, where: Ct<30 indicates occurrence of target Lake Whitefish, Ct≥30 indicates possible occurrence of target; Ct=NA ‘not applicable’ indicates absence of target. Vis ID = consistency of visual identification using a morphological key to species identified by DNA barcoding………………….203

xv Table 6.5. Distribution and abundance of larval Lake Whitefish (Coregonus clupeaformis) observed at Douglas Point, Lake Huron in 2014. Larval abundance and associated standard densities (# larvae 1000m-3 in parentheses). Transect codes (IH,H,B#) are associated with Table 6.1……………………..204

Table 6.6. Axis summary statistics from the Redundancy Analysis of space-time variables on environmental conditions observed at Douglas Point, Lake Huron in 2014. Correlations of environmental conditions and ordination axes. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3. NS=not selected by analysis……..……………………………………….205

Table 6.7. Variance partitioning analysis of the significant space-time factors on environmental conditions at Douglas Point, Lake Huron in 2014, as determined by Redundancy Analysis forward selection. Axis loadings and space-time variable/environmental condition abbreviations correspond to Table 6.3…….206

Table 6.8. Axis summary statistics from the Redundancy Analysis of space- time/environmental conditions on ichthyoplankton standard densities sampled at Douglas Point, Lake Huron in 2014. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3. NS=not selected by analysis.……………………………………………………………………….207

Table 6.9. Variance partitioning analysis of the significant space-time/environment on ichthyoplankton standard densities at Douglas Point, Lake Huron in 2014, as determined by Redundancy Analysis forward selection. Axis loadings and space-time variable/environmental condition abbreviations correspond to Table 6.3……………………………………………………………………………..208

CHAPTER 6: APPENDICES

Table A6.1. Distribution and abundance of non-Lake Whitefish ichythoplankton caught at Douglas Point, Lake Huron in 2014. excluding larval Lake Whitefish. Abundance and standard densities (# larvae 1000m-3 in parentheses) provided. Subtotals provided per week and per region of each of the three bays; Inverhuron Bay, Holmes Bay and Baie du Dore.……………………………..209

xvi Chapter 1. Prologue

Lake Whitefish (Coregonus clupeaformis) ranges across North America and is considered an important cold-water fish species of the Laurentian Great Lakes (Scott &

Crossman 1973). It is well documented that adult Lake Whitefish migrate to rocky spawning shoals where they breed during late autumn and that their embryos overwinter on those shoals relying on the endogenous energy in their yolk (Hart 1930; Cucin &

Regier 1966; Mueller et al. 2015). Embryos typically hatch during early spring, approximately when nearshore ice sheets break up, and they are carried away from the shoals by water currents (Ebener et al. 2008; Lynch et al. 2015). Free embryos are negatively buoyant and must swim constantly to remain suspended in the water column as they search for exogenous food during their ‘critical period’ (Tait 1960; Hoagman

1974; Pothoven et al. 2014). There have been several reports that larval Lake Whitefish actively seek shallow waters along the shoreline, leading to the idea that nearshore nursery habitats are vital to early life history in this species (Faber 1970; Reckahn 1970;

McKenna & Johnson 2009). Due in part to the difficulty tracking free embryos and larvae from the natal shoals (Holmes & Noakes 2002), there have been few investigations designed to test the actual importance of nursery embayments for determining distribution and abundance of larval Lake Whitefish (Freeberg et al. 1990; Roseman et al.

2007; Roseman et al. 2012; Ryan & Crawford 2014).

Currently, there are many outstanding ecological and methodological uncertainties that need to be addressed to develop a reasonable understanding of larval

Lake Whitefish distribution and abundance, including: (a) What are the effects of environmental conditions (natural and anthropogenic) on zooplankton, the main food

1 source of Lake Whitefish larvae? (b) How can we achieve accurate and precise species identification of Lake Whitefish larvae collected in the wild? and (c) What are the effects of environmental conditions (natural and anthropogenic) on the distribution and abundance of Lake Whitefish larvae in relation to hypothesized nursery habitat?

The case study for this thesis is Douglas Point, Lake Huron, home to the Bruce

Nuclear Generating Station (BNGS), which is North America’s largest nuclear power station and has been the focus of several ecological studies and environmental assessments regarding the effects associated with operation (Figure 1.1, Table 1.1). The

BNGS complex is comprised of two distinct facilities (Bruce A Units 1-4, Bruce B Units

5-8), each of which has its own once-through cooling water system that draws Lake

Huron water from intakes 800-1100 m offshore at water depths of 16-18m (Wismer et al.

1986). Each facility pumps Lake Huron intake water to a common forebay channel, from which four nuclear reactors individually pump water through their own condenser systems that cool internal reactor-generated steam, before transporting the heated effluent into a common discharge back into Lake Huron. At full operational capacity, the combined BNGS cooling water system intakes and outputs approximately 360 m3 s-1 of

Lake Huron water heated to 11+°C above ambient conditions; equivalent to 10x peak spring flow of the Saugeen River, one of the largest tributaries to Lake Huron, located approximately 30km to the north (Griffiths 1974; Moore 1976; ESG & BEAK 2000). The environmental and ecological effects of BNGS on Lake Whitefish have come under closer scrutiny over the past decade, largely due to the direct involvement of the Saugeen

Ojibway Nation (collectively the Chippewas of Nawash Unceded First Nation and

Saugeen First Nation). These First Nations pressed for recognition of this species as a

2 ‘Valued Ecosystem Component’ in the Federal Environmental Assessment (EA) process, due to the species cultural, social and economic importance for their communities (SON

2013). Currently, the Saugeen Ojibway Nation deploys the largest Aboriginal commercial fishery on the Great Lakes, and the Main Basin Lake Whitefish population(s) supports the bulk of their harvest (Crawford et al. 2001). In 2003 the Canadian Nuclear Safety

Commission required the development and implementation of a Lake Whitefish Follow-

Up Monitoring Program to address the EA concerns of the First Nations (CNSC 2003). In

2009 the Saugeen Ojibway First Nation and Bruce Power Limited (current lessee and operator of the BNGS facilities) entered into a bi-lateral collaborative research agreement to investigate the ecological effects of BNGS on Lake Whitefish (Crawford et al. 2010).

This thesis is one major component of that research program, specifically focusing on the effect of environmental conditions on distribution and abundance of larval Lake

Whitefish at Douglas Point, Lake Huron, in the following five substantive chapters. The overall goal for this thesis is to investigate potential relationships between environmental conditions and the ecology of larval Lake Whitefish at Douglas Point, Lake Huron by addressing both outstanding key ecological uncertainties and develop methods to meet monitoring needs. In order to achieve this thesis goal, it will be necessary to satisfy a series of specific objectives, organized in the following chapters:

Chapter 2 will investigate the relationship between environmental conditions and plankton, specifically by examining the key ecological uncertainty of which environmental conditions have a major effect on the distribution and abundance of zooplankton assemblages. Food availability has long been recognized as one of the most important factors in determining survival and growth of Lake Whitefish, and it has been

3 hypothesized that free embryos must encounter prey within a critical period of 10-12 days or die of starvation (Freeberg et al. 1990). Zooplankton are known to be an important food source for larval Lake Whitefish (Johnson et al. 2009; Claramunt et al.

2010; Pothoven et al. 2014), and therefore the distribution and abundance of zooplankton assemblages is an essential factor in the ecology of larval Lake Whitefish (Faber 1970).

The explicit goal of Chapter 2 will be to investigate the effects of environmental conditions on the distribution and abundance of zooplankton at Douglas Point, Lake

Huron. Specifically, the relationship between environmental conditions and plankton distribution/abundance will be addressed by testing predictions generated by the following hypotheses, each of which is associated with one of the major sample design factors:

• Seasonal Nutrient Depletion Hypothesis (week-temporal)

• Tributary Nutrient Loading Hypothesis (transect-spatial)

• Forced Displacement Hypothesis (station depth-spatial)

• Diurnal Migration Hypothesis (sample depth-spatial)

The work in this chapter supports inferences regarding the effect of environmental conditions on the location of Lake Whitefish prey over time and space (transect, distance from shore and depth in water column) which presumably drives spatial occurrence of larval Lake Whitefish.

Chapter 3 addresses one of the greatest operational limitations of field work in larval ecology – namely, the accurate and precise species identification of organisms that have not yet reached a definitive phenotype (Richardson et al. 2007; Becker et al. 2015).

This limitation exists because of (a) a general lack of clear morphological and meristic

4 characteristics during the embryo and larval periods (Richardson et al. 2007); (b) sheer volume and diversity of larvae sampled in ichthyoplankton field studies (Durand et al.

2010); (c) the potential for sample degradation during sampling (Ager et al. 2006) and (d) the extremely small size of larval specimens (Becker et al. 2015). The explicit goal of

Chapter 3 will be to evaluate and explore the consistency between visual identification and DNA barcoding identification methods using a case study of larval fishes caught in plankton tows at Stokes Bay, Lake Huron. Specifically, the relative effectiveness of visual and genetic identifications will be evaluated in a structured manner to explicitly tests predictions generated by the following hypotheses:

• Node Complexity Hypothesis (visual)

• Taxonomic Relatedness Hypothesis (visual, genetic)

• Hybridization Hypothesis (visual, genetic)

By quantitatively assessing the accuracy/precision of visual and genetic methods of species identification, this work leads ultimately to a consideration of implementing mixed methods to increase the reliability of interpreting the field distribution/abundance of larval fishes generally, and of larval Lake Whitefish specifically.

Chapter 4 explores the genetic variation that may exist in Lake Huron for spawning-phase Lake Whitefish sampled from a wide variety of locations across all three basins of Lake Huron. Genetic diversity of Lake Whitefish in Lake Huron can be affected by natural geographic and bathymetric separation of the three major basins

(Sloss & Saylor 1975; Bohm 1985) and can result in patterns of variation in morphology, genetics, life history characteristics and behavior (Holmes et al. 2002; Mee et al. 2015).

This investigation makes use of tissue samples from a comprehensive 2012 lake-wide

5 collection of spawning-phase Lake Whitefish from a total of 28 locations around all three basins of Lake Huron, currently the largest single-season collection of any species in

Lake Huron and the largest collection of Lake Whitefish for any water body in North

America. The explicit goal of Chapter 4 will be to investigate basin-level haplotype variation and potential cryptic diversity of Lake Whitefish in Lake Huron by expanding the barcode library across known spawning sites and times. Specifically, the lineages of

Lake Whitefish from the lake-wide sampling program will be evaluated by testing predictions generated from the following hypotheses:

• Panmixis Hypothesis (no isolation)

• Basin Segregation Hypothesis (spatial isolation)

• Sympatric Spawning Hypothesis (temporal isolation)

It should be noted that this chapter is a necessary first step in the development and execution of Chapter 5 (see below).

Chapter 5 determines if a Real-time polymerase chain reaction (qPCR) genetic primer/probe assay can be designed to identify Lake Whitefish using the DNA barcoding region of the target species (Bustin 2005; Rasmussen Hellberg et al. 2011; Gleason &

Burton 2011). The successful development of this assay will depend on whether there is sufficient difference between the DNA barcodes of species found in the Lake Whitefish species complex to design a species-specific probe (Mee et al. 2015). This chapter builds on the haplotype analysis presented in Chapter 4 and the associated increase in Lake

Whitefish DNA sequence quality and quantity publically available on the Barcode of Life

Data System (BOLD; Ratnasingham & Hebert 2007). Real-time PCR has recently been introduced to complement other traditional species identification methods and allows for

6 quantification of biomass for the target species (Galluzzi et al. 2004; Smith & Osborn

2008). The explicit goal of Chapter 5 will be to determine if a qPCR genetic primer/probe assay can be designed to identify Lake Whitefish using the DNA barcode region of the target species. If successful, this method could decrease technical requirements and cost of analysis. Specifically, in the context of this thesis, the development of a functional Lake Whitefish qPCR assay could allow for rapid and cost- effective identification of the target species from unknown larval samples collected in the wild for ecological research and/or environmental assessment.

Chapter 6 examines the key ecological uncertainty of which environmental conditions have a major effect on the distribution and abundance of larval Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron. Sheltered embayments can create important nursery habitats for ichthyoplankton (Klumb et al. 2003), and it has been hypothesized that embayments are crucial for the survival of larval Lake Whitefish

(Holmes et al. 2002; McKenna & Johnson 2009). As part of a previous environmental assessment regarding the effects of BNGS on larval Lake Whitefish, Holmes and Noakes

(2002) explicitly recommended a comparison of larval ecology in the embayments immediately south and north of Douglas Point. The goal of Chapter 6 will be to determine the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron.

Specifically, this chapter will test the predictions generated by a set of hypotheses, each of which will be associated with one of the major sample design factors:

• Larval Pulse Hypothesis (week-temporal)

• Longshore Transport Hypothesis (embayment-spatial)

7 • Embayment Nursery Hypothesis (transect-spatial)

The research in this chapter will allow inferences on the spatio-temporal patterns (if any) observed in distribution/abundance of larval Lake Whitefish in the Douglas Point embayments and help to identify the major environmental factors that could be responsible for those patterns. Ultimately, this chapter seeks to satisfy the overall thesis goal which is to identify the environmental conditions that have a major effect on the ecology of larval Lake Whitefish at Douglas Point, Lake Huron. By presenting the chapters in this structure, this thesis seeks to provide a cohesive story that demonstrates how to apply hypothesis-driven science to address a key ecological uncertainty in Great

Lakes larval fish ecology, using the environmental assessment of North America’s largest nuclear power plant as a case study.

8 1.2. References

Ager D.D., Cord I., Patrick P.H. 2006. Entrainment sampling at Darlington Nuclear Generating Station – 2004. Report Number NK38-REP-07264-10001-R00.

Becker R. A., Sales N. G., Santos G. M., Santos G. B., Carvalho D. C. 2015. DNA barcoding and morphological identification of neotropical ichthyoplankton from the Upper Parana and Sao Francisco. Journal of Fish Biology. doi. 10.1111/jfb.12707.

Bohm E.U. 1985. Lake Huron Nearshore Currents in the Vicinity of Bruce NPD 1983- 1984, Geotechnical and Hydraulic Engineering Department Report No 85301, 1-54.

Brown T. 2007. Whitefish Investigations 2006 Summary. B-REP-00531-00017. Bruce Power. 7pp.

Bustin S.A. 2005. Real-Time PCR. Encyclopedia of Diagnostic Genomics and Proteomics. University of London, London, U.K. 1117-1125.

Claramunt, R.M.,Muir, A.M., Johnson, J., Sutton, T.M., 2010. Spatio-temporal trends in the food habits of age-0 lake whitefish. Journal of Great Lakes Research. 36, 66–72.

CNSC. 2003. Record of proceedings, including reasons for decision in the matter of Bruce Power Inc. Environmental Assessment Screening Report for the return to service of Units 3 & 4 of the Bruce Nuclear Generating Station (NGS) ‘A.’ Canadian Nuclear Safety Commission, Ottawa, Ontario, Canada. 06 January 2003. 25+pp.

Crawford S.S., Muir A., McCann K. 2001. Ecological basis for recommendation of 2001 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Report prepared for the Chippewas of Nawash First Nation, Wiarton, Ontario. 12 October 2001 (revised with references 11 July 2002; revised with response to OMNR comments 02 January 2003). 156+pp.

Crawford S., Boreham D., Fietsch C.-L. 2010. Saugeen Ojibway Nation-Bruce Power (SON BP) Collaborative Lake Whitefish Research Proposal, Proposal prepared by University of Guelph, Guelph, Ontario, McMaster University, Hamilton, Ontario and Bruce Power, Tiverton, Ontario, Canada. 08 April 2010. 17pp.

Cucin D., Regier H. A. 1966. Dynamics and exploitation of lake whitefish in southern Georgian Bay. Journal of Fisheries Research Board of Canada 23(2): 221-274.

Durand J. D., Diatta M. A., Diop K., Trape S. 2010. Multiplex 16S rRNA haplotype- specific PCR, a rapid and convenient method for fish species identification: an application to West African Clupeiform larvae. Molecular Ecology Resources 10: 568- 572.

Ebener M. P., Kinnunen R. E., Schneeberger P. J., Mohr L. C., Hoyle J. A., Peeters P.

9 2008. Management of commercial fisheries for lake whitefish in the Laurentian Great Lakes of North America in: Schechter M. G., Leonard N. J., Taylor W. W. (eds.) International Governance of Fisheries Ecosystems: Learning from the Past, Finding Solutions for the Future. American Fisheries Society Press, Bethesda, Maryland, pp. 99–143.

ESG and Beak Internation Inc. 2000. Bruce Nuclear Power Development Ecological Effects Review Revision R00. Report G99242. 387p.

Faber D. J. 1970. Ecological observations on newly hatched lake whitefish in South Bay, Lake Huron. Lindsey and Woods (eds). In Biology of Coregonid Fishes. 18pp.

Freeberg M. H., Taylor W. W., Brown R. W. 1990. Effect of egg and larval survival on year-class strength of Lake Whitefish in Grand Traverse, Lake Michigan. Transactions of the American Fisheries Society 119: 92-100.

Galluzzi L., Penna A., Bertozzini E., Vila M., Garces E., Magnani M. 2004. Development of a Real-Time PCR assay for rapid detection and quantification of Alexandrium minutum (a Dinoflagellate) Applied Environmenatl Microbiology 70(2): 1199-1206.

Gleason L.U., Burton R.S. 2011. High-throughput molecular identification of fish eggs using multiplex suspension bead arrays. Molecular Ecology Resources, 12,1, 57-66.

Golder Associates Ltd. 2007a. Proposed Work Plans Bruce A Refurbishment for Life Extension Follow-Up Monitoring Program. Bruce Power Limited Report No. 07-1112- 0014. 84pp.

Golder Associates Ltd. 2007b. Whitefish mark/recapture study 2006 Bruce Power. Bruce Power Limited Report No. 05-1112-082. 42pp.

Golder Associated Ltd. 2008d. Bruce New Nuclear Power Plant Project Environmental Assessment. EIS Studies. Aquatic Environment Technical Support Document May 2008. 386 pp.

Golder Associated Ltd. 2008f. Whitefish Emergent Larval Trap and Tow Study 2006 Bruce Power. Bruce Power Limited Report No. 05-1112-070. 34pp.

Griffiths, J. S. 1974. Aquatic Biological Studies – Bruce Nuclear Power Development 1973. Ontario Hydro Research Division Report No. 74-449-H. 30p

Hart J. L. 1930. The spawning and early life history of the whitefish, Coregonus clupeaformis (Mitchell), in the Bay of Quinte, Ontario. Contribut. Can. Biol. Fish NS 6(7):165-214

10 Hoagman W. J. 1974. Vital activity parameters as related to the early life history of larval and post-larval lake whitefish (Coregonus clupeaformis). Blaxter, J. H. S. (eds) in The early life history of fish. Springer-Verlag, New York, NY.

Holmes J. A., Noakes D. L. G. 2002. Whitefish Interactions with Nuclear Generating Station: Recommendations for Monitoring the Effects of the Bruce Nuclear Power Development on Whitefish in Lake Huron. Report. 173pp.

Holmes J. A., Noakes D. L. G., Crawford S. S., Wismer D. A. 2002. Whitefish Interactions with Nuclear Generating Stations. Lake whitefish and round whitefish biology: a review of ecological factors affecting growth, survival, and reproduction. Report. 244pp.

Johnson J. H., McKenna J. E., Chalupnicki M. A., Wallbridge T., Chiavelli R. 2009. Feeding ecology of lake whitefish larvae in Eastern Lake Ontario. Journal of Great Lakes Research 35(4): 603-607.

Klumb R. A., Rudstam L. G., Mills E. L. Schneider C. P., Sawyko P. M. 2003. Importance of Lake Ontario embayments and nearshore habitats as nurseries for larval fishes with emphasis on Alewife (Alosa pseudoharengus). Journal of Great Lakes Research 29(1): 181-198.

Lynch A. J., Taylor W. W., Bear T. D. Jr., Lofgren B. M. 2015. Climate change projections for lake whitefish (Coregonus clupeaformis) recruitment in the 1836 Treaty Waters of the Upper Great Lakes. Journal of Great Lakes Research 41: 415-422.

McKenna Jr. J. E., Johnson J. H. 2009. Spatial and temporal variation in distribution of larval lake whitefish in eastern Lake Ontario: Signs of recovery? Journal of Great Lakes Research 35: 94-100.

Mills, K. H. 1985. Responses of lake whitefish (Coregonus clupeaformis) to fertilization of Lake 226, the Experimental Lakes Area. Canadian Journal of Fisheries and Aquatic Science 42: 129-138.

Mee J. A., Bernatchez L., Reist J. D., Rogers S. M., Taylor E. B. 2015. Identifying designatable units for intraspecific conservation prioritization: a hierarchical approach applied to the lake whitefish species complex (Coregonus spp.) Evolutionary Applications 8: 423-441.

Moore L. P. 1976. 1975 Aquatic Biological Studies Bruce Nuclear Power Development. Ontario Hydro Research Division Report Report No. 76-343-K. 27p.

Mueller C. A., Eme J., Manzon R. G., Somers C. M., Boreham D. R., Wilson J. Y. 2015. Embryonic critical windows: changes in incubation temperature alter survival, hatchling phenotype, and cost of development in lake whitefish (Coregonus clupeaformis) Journal of Comparative Physiology B. 185: 315-331.

11 Pothoven S. A., Hook T. O., Roswell C. R. 2014. Feeding ecology of age-0 lake whitefish in Saginaw Bay, Lake Huron. Journal of Great Lakes Research Supplement 40: 148-155.

Rasmussen Hellberg R.S., Morrissey M.T. 2011. Advances in DNA-Based techniques for the detection of seafood species substitution on the commercial market. Journal of Laboratory Automation. 16: 308-322.

Ratnasingham S., Hebert P. D. N. 2007. BOLD: The Barcode of Life Data System (www.barcodinglife.org). Molecular Ecology Notes 7: 355-364.

Reckahn J. A. 1970. Ecology of young lake whitefish (Coregonus clupeaformis) in South Bay, Manitoulin Island, Lake Huron. Biology of Coregonid Fishes. C. C. Lindsey and C. S. Woods.

Richardson D. E., Vanwye J. D., Exum A. M., Cowen R. K., Crawford D. L. 2007. High- throughput species identification: from DNA isolation to bioinformatics. Molecular Ecology Notes 7: 199-207.

Roseman E. F., Kennedy G. W., Boase J., Manny B. A., Todd T. N., Stott W. 2007. Evidence of Lake Whitefish spawning in the Detroit River: implications for habitat and population recovery. Journal of Great Lakes Research 33: 397–406.

Roseman E. F., Kennedy G., Manny B. A., Boase J., McFee J. 2012. Life history characteristics of a recovering lake whitefish Coregonus clupeaformis stock in the Detroit River, North America. Advances in Limnology 63: 477-501.

Scott W.B., Crossman E.J. 1973. Freshwater Fishes of Canada. Fisheries Research Board of Canada, Ottawa. 966 pp.

Sloss P. W., Saylor J. H. 1975. Measurements of current flow during summer in Lake Huron. NOAA Technical Report ERL 353-GLERL (GLERL Contribution No. 56). Report 45pp.

Smith C. J., Osborn A. M. 2008. Advantages and limiations of quantitative PCR (qPCR) approaches in microbial ecology. FEMS Microbiol Ecology 67: 6-20.

SON. 2013. Oral intervention from Saugeen Ojibway Nation in the matter of Ontario Power Generation Inc., Proposed Environmental Impact Statement for OPG’s Deep Geological Repository (DGR) Project for Low and Intermediate Level Waste. pp. 101, Report prepared for Deep Geologic Repository Joint Review Panel, Canadian Environmental Assessment Agency, Ottawa, Ontario, Canada by Saugeen Ojibway Nation.

Stanley D. 2004. Assessment of the presence and abundance of larval whitefish near Douglas Point, Lake Huron. Stantec Consulting Ltd. Report No. 63123055.2. 34pp.

12 Tait J.S. 1960. The first filling of the swim bladder of salmonoids. Canadian Journal of Zoology 38: 179-187.

Wismer D. A., Hoffer A. D., Chubbuck D. A. 1986. Bruce B Generating Station Pre- Operational Environmental Summary 1975-1983. Environmental Studies and Assessments Department Report No. 86246. 95pp.

13

Figure 1.1. Location of Douglas Point within Lake Huron (black square and inlay). Map shows location of Bruce Nuclear Generating Station sites A and B along with intakes (stars) and discharges (arrows) for each station.

14 Table 1.1. Historical field sampling effort and associated distribution/abundance of larval Lake Whitefish (Coregonus clupeaformis) at Douglas Point, Lake Huron from 1979-2007 excluding forebay entrainment sampling at the Bruce Nuclear Generating Station. Bay= within (In) embayment or outside (Out) of embayment. Offshore = location to shore as either near or far, where nearshore is <3m and offshore is >4m. Orient = orientation of transects as either parallel or perpendicular to shore. Depth = depth contours transects follows (m). Gear= gear used to sample ichthyoplankton with associated effort as either number of tows or hours for larval traps. All=all species of ichthyoplankton sampled. LW=larval lake whitefish sampled. LW-SD= standard density of larval Lake Whitefish (fish/1000m3) sampled. Single data entries in a cell apply to all Region/Bay combinations. Question marks represent missing data. NA= not applicable to sampling regime.

Year Location Sampling Ichthyoplankton Ref (Month) Region Bay Offshore Orient Depth Gear Effort All LW LW- (m) SD 1979 Lagoon Bay Out Wismer (May- Bruce B Discharge Out Plankton et al. Aug) Bruce B Intake Out Near/ Parallel 3, 7, 12 net 126 1500 0 0 1986 South Boundary Pt Out Far (0.5m, tows Holmes Bay In/Out 363 um Gunn Point Out mesh)

1980 Holmes Bay In/Out Plankton Wismer (May- Inverhuron Bay In/Out Near Parallel 1 net 48 1015 0 0 et al. June) (0.5m, tows 1986 363 um mesh) 1981 Lagoon Bay Out Wismer (May- Bruce B Discharge Out Plankton et al. July) Bruce B Intake Out Near/Far Parallel 1, 3, 7 net 126 6983 0 0 1986 South Boundary Pt Out (0.5m, tows Holme Bay In/Out 363 um Gunn Point Out mesh)

15 1982 Lagoon Bay Out Plankton Wismer (May) Bruce B Discharge Out net et al. Bruce B Intake Out Near Parallel 1 (0.5m, 56 2802 0 0 1986 South Boundary Pt Out 363 um tows Holmes Bay In/Out mesh) Gunn Point Out Inverhuron Bay In/Out 2000 Inverhuron In Near/Far ? 1-4 Holmes (May) Baie du Dore In Near/Far ? 1-4 Plankton ? ? 6 ? &Noakes net 2002 (120um, 0.5m) 2001 Inverhuron Bay In 14 10 126 Holmes (Apr- Holmes Bay In Plankton tow 1 13 &Noakes Jun) Douglas Point Out Near/Far ? 1-4 net 4 ? 0 0 2002 MacPherson Bay In (120um, ? 0 0 Baie du Dore In 0.5m) ? 5 63 24 tow 2004 Inverhuron Bay In Near ? Plankton Stanley (Apr) Baie du Dore In/out Near/far ? 1.5-7.1 net 58 35 34 2 2004 (bongo;0 tow .5m; 505um) 2006 Loscombe Bank ? Brown (Apr- Scougall Bank Out Near NA Emergen 13 50 15 ? 2007 May) Welsh Bank t trap trap

2006 Loscombe Bank Out Near ? ? bongo ? ? 11 ? Brown

16 (Apr- Scougall Bank larval 2007 May) Welsh Bank tow 2006 Gunn Point Out Near NA ? Emergen ? ? 1 ? Golder t trap 2007a 2006 Gunn Point Out Near Golder (Mar- Loscombe Bank Out Far 153.3 2008f May) Scougall Bank Out Far ? 3.1-8.2 Larval hrs 39 15 6.61 Welsh Bank Out Far traps

Gunn Point Out Near Paired Golder 2006 Loscombe Bank Out Far Bongo 50 ? 7 1.7 2008f (Mar – Scougall Bank Out Far nets tow May) Welsh Bank Out Far (0.5m; Bruce A intake 505um)

2007 MacPherson Bay Out Far 3-8 Golder (Apr- Douglas Point Out Far ? Larval 17 ? 0 ? 2008d May) Scougall Bank Out Far traps

2007 MacPherson Bay Out Far Both 0.5-20 Paired 52 1 Golder (Apr- Douglas Point Out Far bongo ? 5 ? 2008d May) Scougall Bank Out Far nets 0

17 Chapter 2. Effect of environmental conditions on the distribution and abundance of zooplankton at Douglas Point, Lake Huron

2.1. Abstract

Plankton assemblages serve as Lake Huron's energetic foundation and indicator of water quality. Lake Huron plankton ecology remains largely unexplored, relative to the extensive work that has been done on Lakes Michigan, Erie and Ontario. The goal of this study was to investigate the effects of environmental conditions on the distribution and abundance of zooplankton at Douglas Point, Lake Huron. Zooplankton samples and environmental data were collected from June–July 2013. Zooplankton sampling revealed relatively high frequencies of zooplankton offshore (20, 40m depth contours) and deeper in the water column (15m). A Redundancy Analysis using forward selection indicated that week (time) and sample depth (space) were the most significant space-time variables explaining zooplankton frequency by size. Overall, there was little relationship between environmental variables (e.g. dissolved oxygen) and the distribution and abundance of zooplankton. The results of this investigation support the Diurnal Migration Hypothesis that zooplankton would exhibit a cycle of daily vertical migration to deeper waters.

Strategic and tactical recommendations are provided to modify current plankton sampling monitoring programs for federal environmental assessment of the Bruce Nuclear

Generating Station at Douglas Point, specifically with regard to plankton entrainment by the cooling water intake system.

18 2.2. Introduction

Lake Huron has a diverse plankton community including phytoplankton (e.g., diatoms, green, blue-green, golden algae) and zooplankton (e.g., rotifers, copepods), as well as larval fishes (ichthyoplankton). In general, offshore plankton communities in oligotrophic offshore waters of the lake tend to be less diverse and abundant than nearshore meso/eutrophic waters (Stoermer & Kreis 1980). To date, the majority of Lake

Huron plankton research has focused on the environmental remediation of Saginaw Bay

(see Fishman et al. 2010; Lavrentyev et al. 2014, and references therein); however, there have also been investigations on plankton dynamics elsewhere in the Main Basin

(Barbiero et al. 2009; Bunnell et al. 2011; Barbiero et al. 2013), in the North Channel

(Barton & Griffiths 1984; Munawar et al. 1988; Sprules et al. 1988a) and in Georgian

Bay (Carter & Watson 1977; Ross & Munawar 1988; Sly & Munawar 1988). Despite these previous investigations, Lake Huron plankton ecology remains largely unexplored - relative to the extensive work that has been done on Lakes Michigan, Erie and Ontario

(Munawar et al. 1995; Bunnell et al. 2011). Over the past decade there has been mounting evidence of major ecological changes occurring in Lake Huron (Barbiero et al.

2009; Nalepa et al. 2009; Bunnell et al. 2011; Gobin et al. 2015). In order to understand the new ecological realities of this complex ecosystem, it will be necessary to re-focus attention on the structure and function of the plankton assemblages that serve as Lake

Huron's energetic foundation (Vollenweider et al. 1974).

There are a variety of environmental factors that combine to determine the distribution and abundance of Lake Huron plankton assemblages, including but not limited to: basin-level geostrophic water currents (Balesic & Kwik 1991), wind-induced

19 surface water currents (Hoagman 1973; Balesic & Martin 1987; Balesic & Kwik 1991), nutrient availability (Fishman et al. 2010); light level (Forward 1976; Forward 1988;

Lampert 1989), temperature and ice cover (Weyhenmeyer et al. 1999; Assel et al. 2003).

Depending on the local condition of these environmental factors, Lake Huron phytoplankton and zooplankton can exhibit high levels of variation across different spatio-temporal scales. At a daily scale, plankton often exhibit diurnal vertical migration, moving from surface waters to deeper in the water column (Forward 1976; Forward

1988; Lampert 1989). At a seasonal scale, many ecosystems experience plankton pulses, with species fluctuations (Ellis 1980). At an annual scale, blooms associated with turnover of lakes occur twice yearly depending on the stratification of the lake

(Weyhenmeyer et al. 1999; Peters et al. 2007). It is essential for investigators of Lake

Huron plankton ecology to explicitly consider the importance of these environmental effects when investigating the spatio-temporal patterns observed in plankton assemblage distribution and abundance. Further, the effect of anthropogenic factors must be taken into account when describing plankton ecology.

The Bruce Nuclear Generating Station (BNGS), located at Douglas Point along the southwestern shore of Lake Huron’s Main Basin, is North America’s largest nuclear power plant (Figure 2.1). The BNGS complex is comprised of two distinct facilities

(Bruce A Units 1-4, Bruce B Units 5-8), each of which has its own once-through cooling water system that draws Lake Huron water into a forebay, from which four nuclear reactors individually pump water through their own condenser systems that cool internal reactor-generated steam, before transporting the heated effluent into a common discharge back into Lake Huron (Golder Associates 2011). At full operational capacity, the

20 combined BNGS cooling water system intakes and outputs approximately 360 m3 s-1 of

Lake Huron water heated to 11+°C above ambient conditions; equivalent to 10x peak spring flow of Saugeen River, one of the largest terrestrial tributaries to Lake Huron

(Griffiths 1974; Moore 1976; ESG & BEAK 2000). There are at least five potential mechanisms through which the BNGS cooling water system could affect the distribution and abundance of the plankton community at Douglas Point: (a) intake entrainment of plankton along with Lake Huron source water at Bruce A and B intakes, (b) mechanical/thermal/contaminant mortality of entrained plankton as they are transported through the facilities, (c) output discharge of entrained plankton that have passed through the facilities, (d) output entrainment of plankton in Lake Huron receiving waters that are displaced offshore by the discharge currents, and (e) mechanical/thermal/contaminant mortality of plankton in Lake Huron receiving waters. Taken together, these hypothesized

BNGS cause-effect mechanisms can have a profound influence on the already complex natural structure and function of the Douglas Point plankton assemblage (Holmes &

Noakes 2002; Holmes et al. 2002).

The key ecological uncertainty of this investigation is to what extent do environmental conditions (natural and BNGS) have an effect on the distribution and abundance of zooplankton at Douglas Point, Lake Huron. This uncertainty is important in terms of ecosystem dynamics, fisheries management and environmental assessment.

From an ecosystem perspective, Lake Huron ecosystem dynamics are poorly understood, especially in light of the major changes that are predicted for the Great Lakes generally

(Christie 1974; Magnuson et al. 1990; Nicholls & Hopkins 1993). From a fisheries perspective, the waters surrounding Douglas Point are known as spawning, nursery and

21 feeding grounds for a wide variety of fishes, including Lake Whitefish (Coregonus clupeaformis) which supports the Saugeen Ojibway fishery - the largest and

Constitutionally-protected First Nation commercial fishery deployed on the Great Lakes

(Blair 2000; Crawford et al. 2003). Larvae of Lake Whitefish and other species are completely dependent on local zooplankton as their food source until they advance in development to the point where they can expand their diet to include benthic invertebrates and other fishes (Cucin & Faber 1985; Claramunt et al. 2010; Pothoven et al. 2014). From an environmental assessment perspective, there continues to be a variety of Federal ecological regulatory requirements associated with nuclear facilities at

Douglas Point, including the effects of Bruce Power’s leased operation of BNGS (Bruce

Power 2012) and Ontario Power Generation's proposed construction/operation of a Deep

Geological Repository for Low and Intermediate Level Radioactive Waste (Swanson et al. 2015) on ichthyoplankton. It is important to note that BNGS and associated facilities at Douglas Point exist within the Traditional Territory of the Saugeen Ojibway, and the

Canadian Crown has formally recognized its Duty to Consult the First Nations on these environmental assessments at the highest level of engagement (Newman 2014). Taken together, these factors underscore the importance of understanding how environmental conditions (natural and BNGS) impact the distribution and abundance of zooplankton at

Douglas Point.

For the purpose of this investigation, four qualitatively different cause-effect mechanisms have been explicitly identified as potentially important determinants of zooplankton distribution and abundance at Douglas Point. The Seasonal Nutrient

Depletion Hypothesis states that after spring ice-out and initial phytoplankton blooms,

22 primary production will continue to deplete aquatic nutrients (e.g., phosphorous), resulting in the gradual diminishment of nutrient concentrations and primary production biomass over the remainder of spring and summer (Home & Goldman 1994). The

Tributary Nutrient Loading Hypothesis states that agricultural surface run-off increases nutrient loading in adjacent nearshore waters, which in turn causes a localized increase in primary and secondary biomass (Klumb et al. 2003; McKenna et al. 2008; Leon et al.

2011). The Forced Displacement Hypothesis states that strong wind-induced surface currents will move plankton inshore/offshore in the direction of the prevailing wind

(Evans 1981; Haffner et al. 1984; Evans & Jude 1986). The Diurnal Migration

Hypothesis states that zooplankton will exhibit a cycle of daily vertical migrations to deeper waters in order to avoid negative effects of intense daytime solar radiation in surface waters (Beeton 1960; Geller 1986; Lampert 1989).

The goal of this investigation is to determine which environmental conditions have a major effect on the distribution and abundance of zooplankton at Douglas Point,

Lake Huron. In order to achieve this goal, the following objectives will be satisfied: (1)

Collect water samples and associated environmental data according to a spatio- temporally stratified design at Douglas Point; (2) Evaluate the extent to which space-time variables have an effect on environmental conditions; (3) Evaluate the extent to which space-time variables/environmental conditions have the greatest effect on particle frequency by size (proxy for plankton); and (4) Evaluate the extent to which space-time variables/environmental conditions have the greatest effect on zooplankton frequency by size. In this way, this investigation seeks to explicitly test predictions generated by the

23 four cause-effect hypotheses identified in this study as possible determinants of zooplankton distribution and abundance at Douglas Point, Lake Huron.

2.3. Materials and Methods

Sample Design

A total of 5 latitudinal transects were established in the 2013 study region extending from a nearshore water depth of 3m to an offshore depth of 40m (Figure 2.1). One transect was located specifically to cross over the Bruce B Cooling Water Intake (BI), while another was located specifically to extend offshore from the Bruce B Cooling

Water Outflow (BO). Two reference transects were located south of the Intake transect,

‘upstream’ of the typical northerly alongshore currents that prevail in this region 80% of the time (Wismer et al. 1986): a rocky outcrop at McRae Point (MP) and an embayment at Inverhuron Bay (IH), which are approximately 5km and 1km south of the Intake transect, respectively. One reference transect was located approximately 1km north of the

Outflow transect close to the mid-point of Douglas Point (DP) proper. Sample Stations for each of the five transects were situated at the 3m, 10m, 20m and 40m water depth contours - with the single exception of the Outflow transect which could not be effectively sampled at the 3m depth due to intense outflow at the Bruce B discharge. For each station, on each transect, environmental conditions and water were sampled at sample depths of 1m, 5m and 15m from the surface (where applicable).

Weekly samples were conducted for all transects/stations/sample depths for the five-week period extending from 17 June to 25 July 2013 (Table 2.1). The sequence of sampling transects was randomized for each week, as was the sequence of stations within

24 transects. Specific dates for sampling were dictated by local wind/wave conditions and consideration for safe navigation of 22’ Stanley Pulsecraft workboat (150hp) with a three-person crew.

Environmental Conditions

Environmental conditions were assessed at each station on each transect. A

Furuno BBW GPS (Furuno USA Inc., Camas, Washington, USA) was used to obtain real-time latitude and longitude and used for maintaining vessel location within approximately 20m range of the target station, against variable wind and wave conditions. A RainWise WindLog wind data logger (RainWise Inc., Trenton, ME, USA) was placed at Scotts Point, Baie du Doré (N 44.33551, W 081.55540) and was programmed for 15 minute recordings of average wind speed (km h-1), and average wind direction (° departure from North) which was later converted to a code representing wind destination according to the 8 cardinal compass directions (e.g., N, NE, E, etc.). For statistical analyses, each sample day was associated with average wind speed and modal wind direction for the time period 04:00-12:00. A YSI EXO2 Sonde Probe with handheld readout (Yellow Springs, Inc., Yellow Springs, Ohio, USA) was deployed at the stations/sample depths, in order to assess physico-chemical and primary biomass conditions including: depth (m), temperature (oC), conductivity (µs cm-1), specific conductivity (µs cm-1), salinity (psu), dissolved oxygen (mg L-1), pH, turbidity (FNU), chlorophyll (RFU), chlorophyll (µg L-1), blue-green algae-PC (RFU), blue-green algae-

PC (µg L-1). For each sample depth, the probe was equilibrated for 60s, and then data

25 were recorded for the next 120s and then averaged for a single observation (as recommended by M. Longfield – YSI technical support, pers. comm).

Water Samples

A Rule 2000 electric bilge pump (King Pumps Inc., Miami, Florida, USA) with heavy-duty bilge hose was deployed at the sample stations, in order to collect water samples at each sample depth for assessment of nutrient concentrations and plankton biomass. The bilge pump was lowered to the target depth and operated for 60s to ensure complete flushing of water that had previously been in the bilge hose. For each water sample, pump flow rate was measured by recording time to fill a 9L bucket. A 2L water sample was then collected from the hose and stored in a Nalgene container for subsequent nutrient and chlorophyll analyses. The bilge tube outflow was then placed in the mouth of a 80µm mesh plankton net (35cm diameter, 150cm long; Halltech Environmental Inc.,

Guelph, Ontario, Canada) suspended vertically from a davit; water sampling continued for a period of 8 minutes (approximate equivalent volume of a 5 minute horizontal surface plankton tow). At the end of water sampling, the plankton net was washed down and contents of the cod end were transferred to a storage container with 95% ethanol preservative.

Laboratory Analyses

For each 2L station/depth water sample, two 40mL subsamples were stored in

25x150mm Fisher 14-933D test tubes (Fisher Scientific Co., Ottawa, Ontario, Canada), and sent to the laboratory of Dr. Beatrix Beisner (Université du Quebéc à Montréal-

26 UQAM) for estimation of phosphorus concentration (µg L-1) following the methods described by Wetzel and Likens (1991) and USGS (2003), respectively. A 500 mL water subsample was filtered for chlorophyll using a Pall 4247 150 mL magnetic filter funnel

(Pall Corporation, PortWashington, New York), Whatman GF/F 4.7cm glass microfiber filters (GE Healthcare Life Sciences, Piscataway, New Jersey, USA), and a Rocker 300 oil-free vacuum pump (Rocker Scientific Co., New Taipei City, Taiwan) operating below a maximum pressure of 48kPa. The glass fibre filters were frozen and sent to the UQAM laboratory, where chlorophyll a (µg L-1) was quantified using ethanol extraction and a spectrophotometer (Wintermans & de Mots 1965; Wetzel & Likens 1991).

The preserved plankton samples were sent to UQAM for assessment of particle frequency by size by resuspension/processing with a Brooke Ocean Technology laser optical particle counter (LOPC) (Rolls-Royce Canada - Naval Marine, Dartmouth, Nova

Scotia; Herman et al. 2004) following the methods described by Finlay et al. (2007).

Preserved samples were analyzed using the lab circulator (Brooke Ocean Technology), which consisted of a continuous flow of water through the LOPC at an approximate rate of 0.8 sec L-1. Preserved zooplankton samples were introduced through a chamber and recollected after passage through the LOPC on 75 µm nitex filters. Only particles measured by LOPC between 300-2500µm ESD (estimated spherical diameter) were included in analyses, in order to remove the effects of air bubbles smaller than 300µm

ESD that were frequently present in the lab circulator. Particle size observations were then pooled into 5 non-overlapping size bins (300-500, 501-1000, 1001-1500, 1501-2000,

2001-2500) in 500µm increments, with the exception of the smallest bin that had been truncated to remove air bubbles from consideration. These size bins were created with

27 specific reference to the limiting gape widths that were measured for hatchery-reared larval Lake Whitefish (Coregonus clupeaformis) sampled over eight weeks of post-hatch development (Appendix 2.1).

Finally, the preserved plankton samples were re-assessed visually to estimate percent of zooplankton/non-zooplankton particles in each sample. Each sample was mixed by vigorous shaking and then randomly drawing three 1 mL subsamples, which were placed under a dissecting microscope (Olympus SZ-CTV). Zooplankton, identified on the basis of body segmentation and appendages, were counted, averaged across the three subsamples and then multiplied against corresponding LOPC particle size frequencies to obtain estimates of binned zooplankton size frequencies (300-500, 501-

1000, 1001-1500, 1501-2000, 2001-2500µm).

Statistical Analyses

The distributions of environmental variables were examined visually using boxplots and frequency histograms to assess normality and investigate any outliers. Most variables (temperature, dissolved oxygen, chlorophyll) appeared reasonably normal from histogram distribution plots; pH did not need to be transformed because it was already on a log scale (Nabout et al. 2009). However, phosphorus, turbidity and specific conductivity were clearly non-normal in distribution, and these were transformed using a log(x+1) conversion based on recommendations from the literature (see Muylaert et al.

2000; Gasiunaite et al. 2005; Peng et al. 2012).

Data were imported to CANOCO version 5.0 (Microcomputer Power, Ithaca, USA; ter Braak & Smilauer 2012) in four tables; (a) spatial/temporal data as factors; (b)

28 environmental conditions; (c) particle frequency by size and (d) zooplankton frequency by size (Table 2.2). Redundancy Analysis (RDA) was performed with forward selection of explanatory variables (Legendre &Legendre 2012; Xuemei et al. 2012) for three specific tests:

• effect of space-time variables on environmental conditions;

• effect of space-time variables/environmental conditions on particle size frequency;

and

• effect of space-time variables/environmental conditions on zooplankton size

frequency.

Forward selection allows for the selection of the best subset of environmental conditions to summarize the variation of target response variables over selected explanatory variables. Particle and zooplankton size frequencies were automatically log(x+1) transformed as a recommendation by the CANOCO software. For the purposes of the following analyses, statistical significance was established as p<0.05.

2.4. Results

Effect of Space-Time on Environment

Table 2.3 provides the axis summary statistics for the RDA of space-time variables on environmental conditions sampled at Douglas Point in 2013. Taken together, the factors transect/station depth/week explained 49.4% of the total variation in the environmental data. The first two axes in the RDA were significant and combined to account for 81.1% of the adjusted explained variation of space-time variables on environmental conditions, with Axis 1 explaining 40.9% and Axis 2 explaining 22.8%

29 (Table 2.3). Figure 2.2 shows a biplot of Axes 1 and 2, with space-time explanatory variables depicted as triangles and environmental response variables as vectors. Week features prominently on both axes, with weeks 1 and 3 loading strongly positive on Axis

1, along with high dissolved oxygen/phosphorous and low temperature/conductivity/pH.

Week 2 was notable for being distinctly positive on Axis 2 along with a relatively strong phosphorous loading. The only other space-time explanatory variable to feature prominently in this analysis was sample depth - specifically SDe3 representing samples from the 15m water depth, associated with high dissolved oxygen and low temperature.

Table 2.4 presents variance partitioning from RDA Forward Selection for whole factors of space-time explanatory variables, showing that week (43.8% total variation), sample depth (7.3%) and transect (7.4%) were all significant.

To further investigate the strong effect of week in this analysis, representative time series of sample temperatures were plotted for each water sample depth associated with the 20m station depth on the (a) Inverhuron Bay, (b) Intake and (c) Outflow transects (Figure 2.3). Each of the time series shows a temporary spike in water temperature during week 2, followed by a general increase over weeks 3-5, with stronger effects at the shallower water depths. It should be noted that the 1m and 5m water depth samples tended to be much more similar and warmer at both the Intake and Outflow stations, relative to the Inverhuron station.

Effect of Space-Time/Environment on Particle Size Frequency

Table 2.5 provides the axis summary statistics for the RDA of space- time/environmental conditions on particle size frequency sampled at Douglas Point in

30 2013. Taken together, the combined explanatory factors explained 45.6% of the total variation in the particle size frequency data. Only the first axis in this RDA was significant, and accounted for 89.1% of the adjusted explained variation in particle size frequencies. Figure 2.4 shows a uniplot of Axis 1, with space-time/environmental explanatory variables depicted as triangles, and particle size frequency response variables as vectors. Of the space-time variables with high loadings on Axis 1, sample depth 3

(15m) correlated strongly with particle size frequency, while week 2, sample depth 1

(1m) and station depth 1 (3 m) were all negatively correlated with particle frequency.

Table 2.6 presents variance partitioning from RDA Forward Selection for whole factors of space-time/environmental explanatory variables, showing that sample depth (29.5% total variation) and week (15.4%) were both significant. While this Forward Selection procedure also identified dissolved oxygen and station depth as significant explanatory variables, the associated variation explained was relatively small.

Effect of Space-Time/Environment on Zooplankton Size Frequency

Table 2.7 provides the axis summary statistics for the RDA of space- time/environmental conditions on zooplankton size frequency sampled at Douglas Point in 2013. Taken together, the combined explanatory factors explained 43.2% of the total variation in the zooplankton size frequency data. Once again, only the first axis in this

RDA was significant, and accounted for 98.4% of the adjusted explained variation in particle size frequencies. Figure 2.5 shows a uniplot of Axis 1, with space- time/environmental explanatory variables depicted as triangles, and zooplankton size frequency response variables as vectors. Of all the space-time/environment variables

31 included in this analysis, only week and sample depth were important as explanations of particle size frequencies. Specifically, week 2 and sample depth 1 (1m) were negatively correlated with zooplankton frequencies across all sizes, while sample depth 3 (15m) and week 5 were positively correlated with zooplankton frequencies. Table 2.8 presents variance partitioning from RDA Forward Selection for whole factors of space- time/environmental explanatory variables, showing that sample depth (30.5% total variation), week (15.8%) and dissolved oxygen (3.1%) were significant.

To further investigate the strong effect of week and sample depth in this analysis, representative time series of zooplankton frequencies (all sizes) were plotted for each water sample depth associated with the 20m Station Depth on the (a) Inverhuron Bay, (b)

Intake and (c) Outflow transects (Figure 2.6). Consistent with the Axis 1 uniplot, zooplankton frequencies were consistently low at the 1m water sample depth, and especially during week 2. While zooplankton frequencies appeared somewhat higher toward the end of these time series, the trend was not strong.

It is interesting to note that a single larval fish was collected in one of the 220 bilge pump water samples collected during this study, specifically at the 15m water depth of the 40m station depth on the Douglas Point transect during week 3 sampling. This fish had been mechanically damaged from natural or artificial means (including pump sampling or LOPC processing), and it was not possible to identify the species based on visual characteristics. DNA barcoding was used to genetically identify the specimen to species, which in this case was determined to be Burbot (Lota lota) (Appendix 2.2)*.

* See Chapters 3-6 for DNA barcoding methods

32 2.5. Discussion

Effect of Space-Time on Environment

Forward Selection of the Redundancy Analysis revealed that week, water sample depth and transect were the most significant space-time factors explaining variation of environmental conditions at Douglas Point, Lake Huron in 2013 (Table 2.4). Week 2 was identified as distinct from all other weeks, with a spike in water temperature of approximately 10 C° across all transects (Figure 2.2, Figure 2.3; Appendix 2.3). This spike in water temperature could have been at least partially caused by a minor spring heat wave experienced at the study site during week 2; however the reduction in water temperature observed during week 3 would likely have required an influx of colder offshore water or similar phenomenon. It is interesting to note that water temperature at the 15m stations were substantially lower than the 1 and 5m sample depths during weeks

4 and 5. This phenomenon could be related to the complex hydrodynamic processes associated with stratification and thermocline formation that typically occur during

July/August in Lake Huron nearshore waters (Boyce et al. 1989; Sheng & Rao 2006).

Transect was also identified as a significant space-time factor explaining environmental conditions in this study. Although few consistent differences were observed between the reference transects, the Bruce B Outflow exhibited higher water temperatures at all stations and sample depths - which should come as no surprise, given that this transect was located in the thermal discharge of a major nuclear power plant. In the case of the BNGS, water is discharged into Lake Huron with a ∆T of approximately

11C° relative to ambient water temperatures, with a maximum discharge temperature of

33 32.2°C established for this facility by Environment Canada (OPG 1999; Holmes &

Noakes 2002).

Effect of Space-Time/Environment on Particle Size Frequency

Redundancy Analysis indicated that most space-time variables (sample depth, week, station depth) along with one environmental factor (dissolved oxygen) were significant contributors toward explaining the observed variation in particle frequency

(Table 2.6). Week 2 was identified in a conspicuously negative relationship to particle frequency (Figure 2.4), which could have been related to the water temperature spike during Week 2 as described in the previous subsection. Moore and Folt (1993) noted that heat stress can lead to catastrophic mortality in zooplankton, which could have been reflected in this study as a decrease in particle frequency measured by the LOPC. It remains to be seen if this explanation is supported by the space-time/zooplankton analysis discussed in the next subsection.

Dissolved oxygen showed a positive relationship with particle frequency. It is not clear why dissolved oxygen would be positively correlated with particle frequency, but this may be associated with larger air bubbles (i.e. >300µm) in the water column being counted by the LOPC as 'particles' (Herman 1992; Jensen et al. 2009). It is important to note that the LOPC is unable to distinguish between living particles and detritus, which is why visual zooplankton frequency estimation was included in this investigation (Sprules et al. 1998b; Liebig et al. 2006; Yurista et al. 2009).

34 Effect of Space-Time/Environment on Zooplankton Size Frequency

Redundancy Analysis identified only week, water sample depth and dissolved oxygen as significant factors in explaining variation of zooplankton frequency by size

(Table 2.8). It is important to note that none of the other environmental factors in this study were selected, which is inconsistent with previous investigations of other nearshore

Lake Huron waters. Lavrentyev et al. (2014) found that Chlorophyll A was an important indicator of zooplankton spatial distribution in Saginaw Bay; however, Chlorophyll A was not identified as significant in the 2013 Douglas Point analysis. Week 2 and sample depth 1 (1m) showed strongly negative correlations with zooplankton frequency (Figure

2.5).

Several investigations have concluded that temperature is often the single most important factor structuring zooplankton assemblages (e.g., Moore & Folt 1993; Chen &

Folt 2002). A sudden increase in water temperatures during summer months can cause reductions in zooplankton body size and/or trigger mass mortality events, although these phenomena are very much dependent on species-specific tolerances (Moore & Folt

1993). Chen and Folt (2002) suggest that warmer temperatures can cause distinct population declines of cladocerans and copepods due to an associated decline in body size and fitness. Based on historical sampling, the majority of zooplankton found at

Douglas Point have been cladocerans and copepods (Appendix 2.4), which could account for the noted week 2 zooplankton decline. Ellis (1980) showed that zooplankton at

Douglas Point often exhibit seasonal pulses in abundance, and it is quite possible that the week 2 decline could be one of these pulse events.

35 Hypotheses and Predictions

The goal of this investigation was to determine which environmental conditions have a major effect on the distribution and abundance of zooplankton at Douglas Point,

Lake Huron. In order to achieve this goal, a 2013 field program was executed in a manner that allowed for the empirical evaluation of four hypotheses and associated predictions.

The Seasonal Nutrient Depletion Hypothesis stated that after spring ice-out and initial phytoplankton blooms, primary production would continue to deplete aquatic nutrients (e.g., phosphorous) resulting in the gradual diminishment of nutrient concentrations and primary production biomass over the remainder of spring and summer

(Home & Goldman 1994). If this hypothesis was true, then there would have been an increase in chlorophyll over the first few weeks, followed by a decline (1<2<3>4>5) as zooplankton preyed on phytoplankton and increased in frequency (1=2=3<4<5). The analysis of space-time on environment did identify week as a significant factor on DO, phosphorous, temperature, conductivity and pH; however, Chlorophyll A was not identified as a strong response variable (Figure 2.2). The analysis of space- time/environment on zooplankton also identified week as an important explanatory variable; zooplankton frequencies varied widely across transects and sample depths during the first three weeks; however, there was fairly consistent increase in zooplankton frequencies during weeks 3-5 (Figure 2.5). Thus, while there is definitely a temporal component associated with primary and secondary production in the study region, the phenomenon appears to be much more complex than explained by the Seasonal Nutrient

Depletion Hypothesis. Although findings of this study may not be explained by the

Seasonal Nutrient Depletion Hypothesis, this study is consistent with other research at

36 Douglas Point. Balesic (1983a) noted an alongshore trend in physical and chemical characteristics of the waters surrounding Douglas Point, and he hypothesized that this trend may have been due to the BNGS discharge. Kwik (1982) and Carey (1983) both found that water quality in the study region exhibited greater temporal (seasonal) than spatial variability. These observations are consistent with the findings of this investigation, since week was determined to be the most important space-time factor explaining the variation in environmental conditions.

The Tributary Nutrient Loading Hypothesis stated that agricultural surface run-off increases nutrient loading (nitrogen and phosphorus) in adjacent nearshore waters, which in turn causes a localized increase in primary and secondary production (Klumb et al.

2003; McKenna et al. 2008; Leon et al. 2011). If this hypothesis was true, then there would have been a latitudinal gradient in nutrient concentrations and/or biomass

(Chlorophyll A) from the southerly ‘upstream’ McRae Point transect to a maximum at

Inverhuron Bay with its tributary (Little Sauble River), followed by a decline through dissipation northward to Intake, Outflow and Douglas Point transects

(MPBI>BO>DP). There would also have been an offshore gradient along the

Inverhuron transect in the same response variables from nearshore waters in proximity to the tributary mouth (3m>10m> 20m> 40m). Transect was indeed identified as a significant whole factor in the forward selection component of the space-time on environment analysis; however, it was weak and focused mostly on the Bruce B Intake and Outflow transects rather than on the Inverhuron Bay transect (Table 2.4). Station depth was not identified as a significant explanatory variable in either the space-time on environment analysis or the space-time/environment on zooplankton analysis. Taken

37 together, these results do not support the Tributary Nutrient Loading Hypothesis. It is possible that strong offshore winds and associated upwellings, combined with the longshore current could have resulted in the offshore displacement of nutrients loaded into Inverhuron Bay by its tributary (Balesic & Martin 1987). This is supported by the fact that station depth 4 (40m depth contour) was identified as the most significant in the space-time/environment on particle analysis.

The Forced Displacement Hypothesis stated that strong wind-induced surface currents would move plankton inshore/offshore in the direction of the prevailing wind

(Evans 1981; Haffner et al. 1984; Evans & Jude 1986). If this hypothesis was true, then there would have been an inverse relationship between pre-sampling offshore wind speeds/directions and particle/zooplankton frequency. However, there was no indication that wind speed or wind direction played an important role in explaining either particle or zooplankton frequency. It is interesting to note that the same kind of hypothesis could have applied to offshore particle/zooplankton forcing by the BNGS B Outflow; however, station depth was weak in the space-time/environment analysis on particle size, only explaining 2.5% of variation in the analysis, and inspection of the zooplankton frequencies in Figure 2.6 did not reveal major patterns in either transect or station depth.

Thus, it appears that the Forced Displacement Hypothesis must also be assigned a low probability, based on the evidence and analyses reported here. Leslie (1986) suggested that strong daytime offshore winds may account for nearshore patchiness of particles.

Low particle frequency in nearshore waters may also be attributed to the dynamic water currents of Douglas Point including: a strong undertow in Inverhuron Bay; potential eddies and gyres (Zhao et al. 2012; Howell et al. 2014); changes in river input including

38 nutrients into the system (Pavlac et al. 2012); and the transportation/deposition of sediment by the longshore current which runs from south to north in deeper waters of the lake (Balesic 1983b). It is possible that this investigation missed sampling events when nearshore waters would have higher particle frequencies; upwellings, tributary discharges with strong particle and nutrient inputs, and strong wind events causing resuspension of suspended solids (Makarewicz et al. 2012).

The Diurnal Migration Hypothesis stated that zooplankton would exhibit a cycle of daily vertical migrations to deeper waters in order to avoid negative effects of intense daytime solar radiation in surface waters (Beeton 1960; Geller 1986; Lampert 1989). To support this hypothesis, there would have been a direct relationship between frequency of daytime zooplankton samples and station depths (1m<5m<15m). The analysis of space- time/environment on zooplankton frequency showed that both week and station depth were significant explanatory factors in this study; specifically where SDep3 (15m) was positively correlated with zooplankton frequencies (Figure 2.5, Figure 2.6, Table 2.8).

Ellis (1980) observed that plankton were homogeneously distributed over the entire region surrounding the Bruce A facility, just north of where sampling for the 2013 season took place. Kwik (1982) found no consistent spatial trends in the distribution of either phytoplankton or zooplankton at Douglas Point. The results of this investigation support the Diurnal Migration Hypothesis based on depth distribution. These finding are consistent with many previous observations of diurnal vertically migration by Great

Lakes zooplankton, some of whom travel to great depths (e.g., >15m) during daylight hours (Beeton 1960; Kwik 1982; Geller 1986; Lampert 1989).

39 In conclusion, the results of this investigation have important implications for a variety of different interests and applications. From an ecosystem perspective, this research provides new insight on plankton ecology in a lake ecosystem that is quite poorly understood, especially in the context of major climate-related changes predicted for the Great Lakes generally, and Lake Huron specifically (Magnuson et al. 1990;

Nicholls & Hopkins 1993; Bartolai et al. 2015). Ecologists seem to be quite prepared for major ecosystem changes for Lake Huron in the coming years (Angel & Kunkel 2010;

Kao et al. 2015: Music et al. 2015), but very few have focused on explicit hypotheses/predictions regarding changes in the composition, distribution and abundance of the plankton assemblages around the lake (but see Lynch et al. 2015). Based on the relatively long time series that has already been established for the Douglas Point ecosystem (Wismer et al. 1986; Holmes & Noakes 2002; Stanley 2004), it could serve as a model with which analysts could develop science-based plankton monitoring programs that serve as references for other, even less well-studied, regions of Lake Huron. From a fisheries perspective, this study provides valuable knowledge on the ecology of the zooplankton prey base upon which several species of ichthyoplankton depend for survival

(Wismer et al. 1986; Stanley 2004), including Lake Whitefish which supports the largest harvest of any species in Lake Huron, and the Constitutionally-protected commercial fisheries of the Saugeen Ojibway Nation (Blair 2000; Crawford et al. 2003). To date, the

Lake Whitefish population(s) of Lake Huron have supported relatively large and consistent harvest over the past decades (Lynch et al. 2015) - despite ecological concerns related to major declines in food base (see Appendix 2.5) (Nalepa et al. 2005; Barbiero et al. 2009; Barbiero et al. 2013) and the uncertain effects of accidentally- and intentionally-

40 introduced species (Roseman & Riley 2009; Warner et al. 2015; Kao 2015). From an environmental assessment perspective, there continue to be a variety of Federal regulatory requirements associated with nuclear facilities at Douglas Point, including specific Lake Whitefish monitoring programs associated with Bruce Power’s operation of

BNGS (Bruce Power 2012) and Ontario Power Generation's proposed construction/operation of a Deep Geological Repository for Low and Intermediate Level

Radioactive Waste (Swanson et al. 2015). It would be reasonable to expect that the future of these Federal EAs will place increasing emphasis on the effects of environmental factors (natural and BNGS) on plankton ecology and, in turn, the effects of environmental factors/plankton ecology on the spawning, incubation and larval ecology for the Douglas

Point fish assemblage generally, and for Lake Whitefish specifically. This investigation marks an important shift in conventional EA monitoring effort that has been deployed to date, to a more ecologically-based focus on the important mechanisms that must be considered in a comprehensive, modern environmental assessment (Ebener 1997;

Martinez-Haro et al. 2015; Rigg et al. 2015).

41 2.6. References

Angel J. R., Kunkel K. E. 2010. The response of Great Lakes water levels to future climate scenarios with an emphasis on Lake Michigan-Huron. Journal of Great Lakes Research 36(Supp 2): 51-58.

Assel R., Cronk K., Norton D. 2003. Recent Trends in Laurentian Great Lakes Ice Cover. Climate Change 57: 185-204.

Balesic H. 1983a. Bruce GS ‘B’ Preoperational Biological Study Year III 1981. Report No. 83-451-K. 32 pp.

Balesic H. 1983b. Bruce GS ‘B’ Preoperational Biological Study Year IV 1982-83. Report No. 83-507-K. 42 pp.

Balesic H., Martin I.D. 1987. Bruce GS ‘B’ Preoperational Biological Study Summary, 1979 to 1983. Ontario Hydro Research Division Report N. 87-185-K. 43p.

Balesic H., Kwik J.K. 1991. Bruce GS ‘B’ Commissioning Aquatic Biological Studies Year IV, 1986, and 1983 to 1986 Summary. Ontario Hydro Research Division Report No. 91-63-K. 90p.

Barbiero R. P., Balcer M., Rockwell D. C., Tuchman M. L. 2009. Recent shifts in the crustacean zooplankton community of Lake Huron. Canadian Journal of Fisheries and Aquatic Sciences 66: 816-828.

Barbiero R. P., Nalepa T. F., Lesht B. M., Warren G. J. 2013. Status of phytoplankton, zooplankton and benthos. Pages 10-20 in Riley, S. C., editor. The state of Lake Huron in 2010. Great Lakes Fishery Commission, Ann Arbor, Michigan, Special Publication 13- 01.

Bartolai A. M., He L., Hurst A. E., Mortsch L., Paehlke R., Scavia D. 2015. Climate change as a driver of change in the Great Lakes St. Lawrence River Basin. Journal of Great Lakes Research 41(Supp 1): 45-58.

Barton D. R., Griffiths M. 1984. Benthic invertebrates of the nearshore zone of Eastern Lake Huron, Georgian Bay, and North Channel. Journal of Great Lakes Research 10(4): 407-416.

Beeton A. M. 1960. The vertical migration of Mysis relicta in Lakes Huron and Michigan. Journal of the Fisheries Research Board of Canada 17(4): 517-539.

Blair P. 2000. Taken for "granted": Aboriginal title and public fishing rights in Upper Canada. Ontario History 92: 31-55.

42 Boyce F. M., Donelan M. A., Hamblin P. F., Murthy C., Simons T. 1989. Thermal structure and circulation in the Great Lakes. Atmosphere-Ocean 27: 607-642.

Bruce Power. 2012. Operations Phase Impingement and Entrainment Monitoring Revised Plan, March 2012. Bruce Power, Tiverton, Ontario, Canada.

Bunnell D. B., Davis B. M., Warner D. M., Chriscinske M. A., Roseman E. F. 2011. Planktivory in the changing Lake Huron zooplankton community: Bythotrephes consumption exceeds that of Mysis and fish. Freshwater Biology 56: 1281-1296.

Carey W. E. 1983. Bruce Nuclear Power Development (BNPD) Postoperational Aquatic Survey, Year III -1981. Report No. 83-96-K. 98 pp.

Carter J. C. H., Watson N. H. F. 1977. Seasonal and horizontal distributions of planktonic crustacean in Georgian Bay and North Channel, 1974. Journal of Great Lakes Research 3(1-2): 113-122.

Chen C. Y., Folt C. L. 2002. Ecophysiological responses to warming events by two sympatric zooplankton species. Journal of Plankton Research 24(6): 579-589.

Christie W. J. 1974. Changes in the fish species composition of the Great Lakes. Journal of the Fisheries Research Board of Canada 31(5): 827-854.

Claramunt R. M., Muir A. M., Johnson J., Sutton T. M. 2010. Spatio-temporal trends in the food habitats of age-0 lake whitefish. Journal of Great Lakes Research 36: 66-72.

Crawford S.S. 2001. Salmonine introductions to the Laurentian Great Lakes: an historical review and evaluation of ecological effects. Canadian Special Publication of Fisheries and Aquatic Sciences 132. 205 pp.

Crawford S., McCann K., Muir A. 2003. 2003 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Report prepared for the Chippewas of Nawash Unceded First Nation R.R. #5, Wiarton, Ontario N0H 2T0. 109pp.

Cucin D., Faber D. J. 1985. Early life studies of lake whitefish (Coregonus clupeaformis, Mitchill), cisco (Coregonus artedii) and yellow perch (Perca flavescens) in Lake Opeongo, Ontario. Ontario Fisheries Technical Report Series, 16.

Ebener M.P. 1997. Recovery of lake whitefish populations in the Great Lakes: a story of successful management and just plain luck. Fisheries 22: 18-20.

Ellis J. D. 1980. Preoperational Aquatic Biological Studies Bruce Nuclear Power Development 1973-1975. Report No. 80-291-K. 48 pp.

43 ESG and Beak Internation Inc. 2000. Bruce Nuclear Power Development Ecological Effects Review Revision R00. Report G99242. 387p.

Evans M. S. 1981. Distribution of zooplankton populations within and adjacent to a thermal plume. Canadian Journal of Fisheries and Aquatic Science 38(4): 441-448.

Evans M. S., Jude D. J. 1986. Recent shifts in Daphnia community structure in southeastern Lake Michigan: A comparison of the inshore and offshore region, 12. Limnology and Oceanography 31(1):56-67.

Finlay K., Beisner B. E., Barnett A. J. D. 2007. The use of the laser optical plankton counter to measure zooplankton size, abundance, and biomass in small freshwater lakes. Limnology and Oceanography: Methods 5: 41-49.

Fishman D.B., Adlerstein S. A., Vanderploeg H. A., Fahnenstiel G. L., Scavia D. 2010. Phytoplankton community composition of Sagina Bay, Lake Huron, during the zebra mussel (Dresissena polymorpha) invasion: A multivariatie analysis. Journal of Great Lakes Research 36: 9-19.

Forward R. B. Jr. 1976. Light and diurnal vertical migration: photobehaviour and photophysiology of plankton. Photochemical and Photobiological Review 157-209.

Forward R. B. Jr. 1988. Diel vertical migration: zooplankton photobiology and behavior. Oceanography and Marine Biololy - An Annual Review 26: 361-393.

Gasiunaite Z. R., Cardoso A. C., Heiskanen A-S., Henriksen P., Kauppila P., Olenina I., Pilkaityte R., Purina I., Razinkovas A., Sagert S., Schubert H., Wasmund N. 2005. Seasonality of coastal phytoplankton in the Baltic Sea: Influence of salinity and eutrophication. Estuarine, Coastal and Shelf Science 65: 239-252.

Geller W. 1986. Diurnal vertical migration of zooplankton in a temperate great lake (L. Constance): A starvation avoidance mechanism? Archiv für Hydrobiologie 1 (Suppl. 74): 1-60.

Gobin J., Lester N. P., Cottrill A., Fox M. G., Dunlop E. S. 2015. Trends in growth and recruitment of Lake Huron lake whitefish during a period of ecosystem change, 1985 to 2012. Journal of Great Lakes Research 41: 405-414

Golder Associates. 2008e. Final Report on Bruce A Thermal Impact Monitoring Study CR 284257. Report No. 06-1112-066. 158p.

Golder Associates. 2011. Bruce A refurbishment for life extension and continued operations environmental assessment follow-up program. Draft operations phase impingement and entrainment monitoring plan. Report No. 11-1151-0243. 45p.

44 Griffiths, J. S. 1974. Aquatic Biological Studies – Bruce Nuclear Power Development 1973. Ontario Hydro Research Division Report No. 74-449-H. 30p.

Haffner G. D., Yallop M. L., Hebert P. D. N., Griffiths H. M. 1984. Ecological significance of upwelling events in Lake Ontario. Journal of Great Lakes Research 10(1): 28-37.

Herman A. W. 1992. Design and calibration of a new optical plankton counter capable of sizing small zooplankton. Deep-Sea Research 39(3/4): 395-415.

Herman A. W., Beanlands B., Phillips E. F. 2004. The next generation of optical plankton counter: the Laser-OPC. Journal of Plankton Research 26: 1135-1145.

Hoagman W. J. 1973. The hatching, distribution, abundance, growth, and food of the larval lake whitefish (Coregonus clupeaformis Mitchell) of Central Green Bay, Lake Michigan. Institute of Freshwater Research, Drottningholm. Report 53: 1-20.

Holmes J. A., Noakes D. L. G. 2002. Whitefish Interactions with Nuclear Generating Stations. Recommendations for monitoring the effects of the Bruce Nuclear Power Development on Whitefish in Lake Huron. Report. 173 pp.

Holmes J. A., Noakes D. L. G., Crawford S. S., Wismer D. 2002. Whitefish Interactions with Nuclear Generating Stations. Aquatic Effects of Nuclear Generating Stations on Fishes. Report. 352 pp.

Home A. J., Goldman C. R. 1994. Limnology. New York: McGraw-Hill.

Howell E. T., Barton D. R., Fietsch C-L., Kaltenecker G. 2014. Fine-scale nutrient enrichment and water quality on the rural shores of Southeast Lake Huron. Journal of Great Lakes Research 40: 126-140.

Jensen O. P., Yurista P. M., Hrabik T. R., Stockwell J. D. 2009. Densities and diel vertical migration of Mysis relicta in Lake Superior: a comparison of optical plankton counter and net-based approaches. Verhandlungen des Internationalen Verein Limnologie 30(6): 957-963.

Johnson A. F. 1973. Preliminary environmental study- aquatic biology progress report – Bruce II Site, 1972. The Hydro-electric Power Commission of Ontario Research Division Report. 15pp.

Kao Y-C. 2015. Modeling the effects of climate change, nutrients, and invasive species on Lake Huron food webs. PhD Thesis. University of Michigan.

Kao Y-C. Madenjian C. P., Bunnell D. B., Lofgren B. M., Perroud M. 2015. Potential effects of climate change on the growth of fishes from different thermal guilds in Lake Michigan and Huron. Journal of Great Lakes Research 41(2): 423-435.

45 Klumb R.A., Rudstam L. G., Mills E. L., Schneider C. P., Sawyko P. M. 2003. Importance of Lake Ontario embayments and nearshore habitats as nurseries for larval fishes with emphasis on alewife (Alosa pseudoharengus). Journal of Great Lakes Research 29(1): 181-198.

Kwik J. K. 1982. BNPD Postoperational Survey Year II, 1980. Report No. 82-182-K. 8 pp.

Lampert W. 1989. The adaptive significance of diel vertical migration of zooplankton. Functional Ecology 3: 21-27.

Lavrentyev P. J., Vanderploeg H. A., Franze G., Chacin D. H., Liebig J. R., Johengen T. H. 2014. Microzooplankton distribution, dynamics, and trophic interactions relative to phytoplankton and quagga mussels in Saginaw Bay, Lake Huron. Journal of Great Lakes Research 40 (Supplement 1): 95-105.

Legendre P., Legendre L. 2012. Numerical Ecology 3rd Edition. Amsterdam, The Netherlands.

Leon L. F., Smith R. E. H., Hipsey M. R., Bocaniov S. A., Higgins S. N., Hecky R. E., Antenucci J. P., Imberger J. A., Guildford S. J. 2011. Application of a 3D hydrodynamic- biological model for seasonal and spatial dynamics of water quality and phytoplankton in Lake Erie. Journal of Great Lakes Research 37: 41-53.

Leslie K. J. 1986. Nearshore contagion and sampling of freshwater larval fish. Journal of Plankton Research 8(6): 1137-1147.

Liebig, J.R., Vanderploeg, H.A., Ruberg, S.A., 2006. Factors affecting the performance of the optical plankton counter in large lakes: insights from Lake Michigan and laboratory studies. Journal of Geophysical Research. 111, C05S02.

Lynch A. J., Taylor W. W., Beard T. D. Jr., Lofgren B. M. 2015 Climate change projections for lake whitefish (Coregonus clupeaformis) recruitment in the 1836 Treaty Waters of the Upper Great Lakes. Journal of Great Lakes Research 41(2): 415-422.

Magnuson J. J., Meisner J. D., Hill D. K. 1990. Potential changes in the thermal habitat of Great Lakes fish after global climate warming. Transactions of the American Fisheries Society 119(2): 254-264.

Makarewicz J. C., Lewis T. W., Boyer G. L., Edward W. J. 2012. The influence of streams on nearshore water chemistry, Lake Ontario. Journal of Great Lakes Research 38:62-71.

Martinez-Haro M., Beiras R., Bellas J., Capela R., Coelho J-P., Lopes I., Moreira-Santos M., Reis-Henriques A. M., Ribeiro R., Santos M. M., Marques J-C. A review on the ecological quality status assessment in aquatic systems using community based indicators

46 and ecotoxicological tools: what might be the added value of their combination? Ecological Indicators 48: 8-16.

McKenna J. E. Jr., Davis B. M., Fabrizio M. C., Savino J. F., Todd T. N., Mur M. 2008. Ichthyoplankton assemblages of coastal west-central Lake Erie and associated habitat characteristics. Journal of Great Lakes Research 34: 755-769.

Moore L. P. 1976. 1975 Aquatic Biological Studies Bruce Nuclear Power Development. Ontario Hydro Research Division Report Report No. 76-343-K. 27p.

Moore M., Folt C. 1993. Zooplankton body size and community structure: effects of thermal and toxicant stress. Trends in Ecology and Evolution 8(5): 178-183.

Munawar M., Munawar I. F., McCarthy L. H., Duthie H. C. 1988. Phycological studies in the North Channel, Lake Huron. Hydrobiologia 163: 119-134.

Munawar M., Munawar I. F., Legner M. 1995. Lake Huron, North Channel, and Georgian Bay ecosystems: Part I: Phytoplankton structure and ecology in: The Lake Huron Ecosystems: Ecology, Fisheries and Management pp.39-78 (eds) Munawar, M., T. Edsall and J. Leach. Academic Publishing, Amsterdam.

Music B., Frigon A., Lofgren B., Turcotte R., Cyr J-F. 2015. Present and future Laurentian Great Lakes hydroclimatic conditions as simulated by regional climate models with an emphasis on Lake Michigan-Huron. Climate Change 130(4): 603-618.

Muylaert K., Sabbe K., Vyverman W. 2000. Spatial and temporal dynamics of phytoplankton communities in a freshwater tidal estuary (Schelde, Belgium). Estuarine, Coastal and Shelf Science 50: 673-687.

Nabout J. C., Siqueira T., Bini L. M., Nogueira I de S. 2009. No evidence for environmental and spatial processes in structuring phytoplankton communities. Acta Oecologica 35: 720-726.

Nalepa, T.F., D.L. Fanslow & G. Messick. 2005b. Characteristics and potential causes of declining Diporeia spp. populations in southern Lake Michigan and Saginaw Bay, Lake Huron. pp. 157-188. In: L.C. Mohr & T.F. Nalepa (ed.) Proceedings of a Workshop on the Dynamics of Lake Whitefish (Coregonus clupeaformis) and the Amphipod Diporeia spp. in the Great Lakes, Great Lakes Fishery Commission Technical Report 66. Ann Arbor, Michigan, USA.

Nalepa T. F., Pothoven S. A., Fanslow D. L. 2009. Recent changes in benthic macroinvertebrate populations in Lake Huron and impact on the diet of lake whitefish (Coregonus clupeaformis). Aquatic Ecosystem Health and Management 12(1): 2-10.

Newman D. G. 2014. Revisiting the duty to consult Aboriginal Peoples. Purich Publishing Ltd., Saskatoon, Saskatchewan, Canada. 184pp.

47 Nicholls K. H., Hopkins G. J. 1993. Recent changes in Lake Erie (North Shore) phytoplankton: cumulative impacts of phosphorus loading reductions and the zebra mussel introduction. Journal of Great Lakes Research 19(4): 637-647.

OPG (Ontario Power Generation). 1999. Bruce 5-8 environmental effects report. Ont. Power Gen. - Nuclear, Tech. Sup. Environ. Aff. Subdiv., NK29-REP-07010.011- 0001R00.

Pavlac M. M., Smith T. T., Thomas S. P., Makarewicz J. C., Edwards W. J., Pennuto C. M., Boyer G. L. 2012. Assessment of phytoplankton distribution in the nearshore zone using continuous in situ fluorometry. Journal of Great Lakes Research 38: 78-84.

Peeters F., Straile D., Lorke A., Ollinger D. 2007. Turbulent mixing and phytoplankton spring bloom development in a deep lake. Limnology and Oceanography 52(1): 286-298.

Peng S., Qin X., Shi H., Zhou R., Dai M., Ding D. 2012. Distribution and controlling factors of phytoplankton assemblages in a semi-enclosed bay during spring and summer. Marine Pollution Bulletin 64: 941-948.

Pothoven S. A., Hook T. O., Roswell C. R. 2014. Feeding ecology of age-0 lake whitefish in Saginaw Bay, Lake Huron. Journal of Great Lakes Research Supplement 40: 148-155.

Rigg D. K., Wacksman M. N., Iannuzzi J., Baker T. F., Adams M., Greeley M. S. Jr. 2015. Assessing ecological risks to the fish community from residual coal fly ash in Watts Bar Reservoir, Tennessee. Integrated Environmental Assessment and Management 11(1): 88-101.

Roseman E. F., Riley S. C. 2009. Biomass of deepwater demersal forage fishes in Lake Huron, 1994-2007: Implications for offshore predators. Aquatic Ecosystem Health and Management 12(1): 29-36.

Ross P. E., Munawar M. 1988. Zooplankton filtration rates in Lake Huron, Georgian Bay and North Channel. Hydrobiologia 163: 173-177.

Sheng J., Rao Y. R. 2006. Circulation and thermal structure in Lake Huron and Georgian Bay: Application of a nested-grid hydrodynamic model. Continental Shelf Research 26: 1496-1518.

Sly P. G., Munawar M. 1988. Great Lakes Manitoulin: Georgian Bay and the North Channel. Hydrobiologia 163: 1-19.

Sprules W. G., Munawar M., Jin E. H. 1988a. Plankton community structure and size spectra in the Georgian Bay and North Channel ecosystems. Hydrobiologia 163:135-140.

Sprules, W.G., Jin, E.H., Herman, A.W., Stockwell, J.D., 1998b. Calibration of an optical

48 plankton counter for use in fresh water. Limnology and Oceanography. 43, 726–733.

Stanley D. 2004. Assessment of the presence and abundance of larval lake whitefish near Douglas Point, Lake Huron. Stantec Consulting Ltd. Red. 63123055.2. Report 34pp.

Stoermer E. F., Kreis Jr., R. G. 1980. Phytoplankton composition and abundance in Southern Lake Huron. Report EPA-600/3-80-061. Great Lakes Research Division, University of Michigan, Ann Arbor, Michigan, USA

Swanson, S., J.F. Archibald and G. Muecke. 2015. Joint Review Panel environmental assessment report - Deep Geologic Repository for Low and Intermediate-level Radioactive Waste Project (May 6, 2015). CEAA Reference No. 17520, Canadian Environmental Assessment Agency, Ottawa, Ontario, Canada. 32pp.

Ter Braak, C. J. F., Smilauer P. 2012. Canoco reference manual and user’s guide: software for ordination (version 5.0). Microcomputer Power, Ithaca, NY. USA.

USGS. 2003. Methods of analysis by the U.S. Geological Survey National Water Quality Laboratory - evaluation of alkaline persulfate digestion as an alternative to Kjeldahl digestion for determination of total and dissolved nitrogen and phosphorus in water, Water Resources Investigations Report 03-4174. U.S. Geological Survey, Denver, Colorado, USA

Vollenweider R. A., Munawar M., Stadelman P. 1974. A comparative review of phytoplankton and primary production in the Laurentian Great Lakes. Journal of the Fisheries Research Board of Canada 31: 739-762.

Warner D.M., Lesht B. M. 2015. Relative importance of phosphorus, invasive mussels and climate for patterns in chlorophyll as and primary production in Lakes Michigan and Huron. Freshwater Biology 60(5): 1029-1043.

Wetzel R.G., Likens G. E. 1991. Limnological analysis, 2nd edition. Springer-Verlag. New York, New York, USA.

Weyhenmeyer G. A., Blenckner T., Pettersson K. 1999. Changes of the plankton spring outburst related to the North Atlantic Oscillation. Limnology and Oceanography 44(7): 1788-1792.

Wintermans J.F., de Mots A. 1965. Spectrophotometric characteristics of chlorophylls a and b and their pheophytins in ethanol. Biochimica et Biophysica Acta 109: 448-452.

Wismer D. A., Hoffer A. D., Chubbuck D. A. 1986. Bruce B Generating Station Pre- Operational Environmental Summary 1975-1983. Environmental Studies and Assessments Department Report No. 86246.

49 Xuemei L., Yuhe Y., Tanglin Z., Weisong F., Hongyi A., Qingyun Y. 2012. Seasonal variation of plankton communities influenced by environmental factors in an artificial lake. Chinese Journal of Oceanology and Limnology 30(3): 397-403.

Yurista P. M., Kelly J. R., Miller S. E. 2009. Lake Superior zooplankton biomass: Alternate estimates from a probability-based net survey and spatially extensive LOPC surveys. Journal of Great Lakes Research 35: 337-346.

Zhao J., Rao Y. R., Sheng J. 2012. Numerical study of dispersion and hydrodynamic connectivity of near-surface waters in Lake Huron. Water Quality Research Journal of Canada 47: 3-4.

50

Figure 2.1. Map of Douglas Point, Lake Huron showing the 2013 sampling transects (DP= Douglas Point, BO=Outflow for BNGS ‘B’, BI=intake for BNGS ‘B’, IH=Inverhuron Bay, MP=McRae Point). Circles represent sampling stations at 3m, 10m, 20m and 40m for each transect, with the exception of the discharge (BO) where strong currents prevented effective sampling at the 3m depth. Stars indicate location of cooling water intake for BNGS ‘A’ and ‘B’ facilities. Grey arrows indicate discharge outflows of cooling water systems.

51

Figure 2.2. Biplot of the Redundancy Analysis after a forward selection of the relationship between space-time variables on environmental conditions sampled at Douglas Point, Lake Huron in 2013. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

52 A 25

20

15 1m 10 5m

Temperature (°C ) 5 15m

0 0 1 2 3 4 5 Week

B 25

20

15 1m 10 5m 5 Temperature (°C ) 15m 0 0 1 2 3 4 5 Week C 25

20

15 1m 10 5m

Temperature (°C ) 5 15m

0 0 1 2 3 4 5 Week

Figure 2.3.A-C. Water temperatures (°C) sampled at the 20m station depth for all sample depths (1,5,15m) over five weeks for transects (A) Inverhuron (IH), (B) Intake B (BI) and (C) Outflow B (BO) respectively at Douglas Point, Lake Huron sampled in 2013.

53

Figure 2.4. Uniplot of the Redundancy Analysis after a forward selection of the relationship between space-time and environmental conditions on particle size sampled at Douglas Point, Lake Huron in 2013. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

54

Figure 2.5. Uniplot of the Redundancy Analysis after a forward selection of the relationship between space-time variables and environmental conditions on zooplankton frequency by size sampled at Douglas Point, Lake Huron in 2013. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

55 A 90 80 70 60 50 1m 40 30 5m 20 15m 10 Zooplankton Frequency 0 0 1 2 3 4 5 Week

B 90 80 70 60 50 1m 40 30 5m 20 15m 10 Zooplankton Frequency 0 0 1 2 3 4 5 Week

C 90 80 70 60 50 1m 40 30 5m 20 15m

Zooplankton Frequency 10 0 0 1 2 3 4 5 Week

Figure 2.6. A-C. Zooplankton frequency sampled at the 20m station depth for all sample depths (1,5,15m) over five weeks for transects (A) Inverhuron (IH), (B) Intake B (BI) and (C) Outflow B (BO) respectively at Douglas Point, Lake Huron sampled in 2013.

56

Table 2.1. Dates of plankton samples for each of the five transects at Douglas Point, Lake Huron during the 2013 field season. MP= McRae Point, IH=Inverhuron Bay, BI=Bruce Intake B, BO=Bruce Outflow B, DP=Douglas Point.

Transect Week 1 Week 2 Week 3 Week 4 Week 5 DP 19/20 June 27 June 05 July 09 July 16 July BO 20 June 27 June 05 July 09 July 17 July BI 20 June 25 June 03 July 13 July 17 July IH 17/19 June 26 June 03 July 13 July 16 July MP 18/19 June 26 June 05 July 09 July 16 July

57

Table 2.2. Space-time variables and environmental conditions sampled at Douglas Point, Lake Huron in 2013 with associated codes referenced in Redundancy Analyses.

Type Variable Units Code Spatial Transect 1- McRae Point (MP) 1_MP 2- Inverhuron (IH) 2_IH 3- Bruce Intake B (BI) 3_BI 4- Bruce Output B (BO) 4_BO 5- Douglas Point (DP) 5_DP Station Depth Depth 3 m Dep1 Depth 10 m Dep2 Depth 20 m Dep3 Depth 40 m Dep4 Sample Depth Depth 1 m SDep1 Depth 5 m SDep2 Depth 15 m SDep3 Temporal Week Week 1 Week1 Week 2 Week2 Week 3 Week3 Week 4 Week4 Week 5 Week5 Environmental Temperature °C Temp Specific conductivity µs cm-1 L-Scond (log (x+1)) pH pH Turbidity (log (x+1)) FNU L-Turbid Wind Speed kph WindS8 Wind Direction North No North East NE East E South West SW Phosphorus (log (x+1)) µg L-1 L-P Chlorophyll A µg L-1 Chl A Dissolved oxygen µg L-1 DO

58 Table 2.3. Axis summary statistics from the Redundancy Analysis of space-time variables on environmental conditions sampled at Douglas Point, Lake Huron in 2013. Adjusted explained variation of 49.4%. Correlations of environmental conditions and ordination axes. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

Statistics Axis 1 Axis 2 Environmental Conditions Windir.No -0.1088 -0.5935 Windir.NE -0.1012 0.6633 Windir.E -0.0573 -0.4222 Windir.SW 0.3966 0.2166 WindS8 -0.0045 0.1118 Temp -0.8658 -0.11 L-Scond -0.6757 0.1676 DO 0.6659 0.229 pH -0.7547 0.0787 L-Turbid 0.0012 0.3319 L-P 0.4034 -0.4244 ChlA 0.1175 -0.0969

Axis Summary Eigenvalues 0.2151 0.1196 Pseudo-canonical correlation 0.8951 0.8769 Explained variation (cumulative) 21.51 33.47 Explained fitted variation (cumulative) 40.99 63.78 F 55.31 36.1 p 0.001 0.001

59 Table 2.4. Variation partitioning analysis of the significant space-time factors on environmental conditions in Redundancy Analysis forward selection sampled at Douglas Point, Lake Huron in 2013. Percent (%) Explains is the percent (%) variation explained solely by that individual factor. p values are applied to factor as a whole. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

Factor Factor Axis 1 Axis 2 % F p (whole) Explains Week Week5 -0.6011 0.6846 13.7 33.7 0.001 Week4 -0.244 -0.4363 10.9 30.5 Week2 -0.1673 -0.6089 10.8 35.2 Week3 0.5174 0.3551 4.2 14.5 Week1 0.4839 -0.0102 4.2 14.5 SDepth SDe3 0.368 0.0416 5.1 19.1 0.001 SDe2 0.0174 -0.0257 1.1 4.0 SDe1 -0.3309 -0.0109 1.1 4.0 Transect 3_BI 0.0951 0.2197 2.4 9.7 0.001 2_IH 0.1789 -0.0929 1.8 7.6 4_BO -0.1736 0.0926 1.2 4.9 5_DP -0.0974 -0.0884 1.0 4.1 1_MP -0.0109 -0.1279 1.0 4.1

60 Table 2.5. Axis summary statistics from the Redundancy Analysis of space-time variables and environmental conditions on particle size frequency sampled at Douglas Point, Lake Huron in 2013. Adjusted explained variation is 45.6% (R2=63.9%). Correlations of environmental conditions and ordination axes. NS = not selected by analysis. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

Statistics Axis 1 Axis 2 Environmental Conditions Windir.No NS NS Windir.NE NS NS Windir.E NS NS Windir.SW NS NS WindS8 NS NS Temp NS NS L-Scond NS NS DO 0.6209 -0.0342 pH NS NS L-Turbid NS NS L-P NS NS ChlA NS NS Response variables P0.3-0.5 0.7469 0.0983 P0.5-1.0 0.7067 -0.0037 P1.0-1.5 0.3914 -0.3657 P1.5-2.0 0.1103 -0.1845 P2.0+ 0.1575 -0.1704 Axis Summary Eigenvalues 0.4603 0.0385 Pseudo-canonical correlation 0.7754 0.5607 Explained variation (cumulative) 46.03 49.87 Explained fitted variation (cumulative) 89.06 96.50 F 162.0 14.60 p 0.001 0.367

61 Table 2.6. Variation partitioning analysis of the significant space-time variables and environmental conditions on particle size frequency in Redundancy Analysis forward selection sampled at Douglas Point, Lake Huron in 2013. Percent (%) Explains is the percent (%) variation explained solely by that individual factor. p values are applied to factor as a whole. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

Factor Factor Axis 1 Axis 2 % F p (whole) Explains Sample SDe1 -0.715 0.0003 21.1 57 0.001 Depth SDe2 0.203 0.025 4.2 11.9 SDe3 0.6099 -0.0285 4.2 11.9 Week Week2 -0.4527 0.0966 9.5 30.8 0.001 Week5 0.2339 0.6834 2.4 7.9 Week4 0.2699 -0.3735 1.3 4.6 Week1 -0.0032 -0.1405 1.3 4.6 Week3 -0.062 -0.2568 0.9 2.9 DO DO 0.6209 -0.0342 3.1 11 0.001 Station Dep4 -0.0535 0.419 2.1 7.9 0.001 Depth Dep2 -0.0516 -0.3992 0.4 1.3 Dep1 -0.2149 0.1607 <0.1 0.3 Dep3 0.2273 -0.1613 <0.1 0.3

62 Table 2.7. Axis summary statistics from the Redundancy Analysis of space-time variables and environmental conditions on zooplankton size frequency sampled at Douglas Point, Lake Huron in 2013. Adjusted explained variation is 43.2% (R2=91.3%). Correlations of environmental conditions and ordination axes. NS=not selected by analysis. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

Statistics Axis 1 Axis 2 Environmental Conditions Windir.No NS NS Windir.NE NS NS Windir.E NS NS Windir.SW NS NS WindS8 NS NS Temp NS NS L-Scond NS NS DO 0.6371 -0.3947 pH NS NS L-Turbid NS NS L-P NS NS ChlA NS NS Response variables Z0.3-0.5 0.6693 0.0518 Z0.5-1.0 0.6698 -0.0596 Z1.0-1.5 0.4767 -0.3128 Z1.5-2.0 0.1006 -0.1012 Z2.0+ 0.1607 -0.068 Axis Summary Eigenvalues 0.4875 0.0070 Pseudo-canonical correlation 0.7086 0.6101 Explained variation (cumulative) 48.75 49.45 Explained fitted variation (cumulative) 98.35 99.76 F 181.0 2.600 p 0.001 1.000

63 Table 2.8. Variation partitioning analysis of the significant space-time variables and environmental conditions on zooplankton size frequency in Redundancy Analysis forward selection sampled at Douglas Point, Lake Huron in 2013. Percent (%) Explains is the percent (%) variation explained solely by that individual factor. p values are applied to factor as a whole. Space-time variable and environmental condition abbreviations correspond to those in Table 2.2.

Factor Factor Axis 1 Axis 2 % F p (whole) Explains DO DO 0.6371 -0.3947 3.1 11.5 0.002 Week Week2 -0.4974 -0.2273 11.9 40.3 0.001 Week5 0.3703 0.7519 3.3 11.7 Week3 -0.0193 -0.1832 0.2 0.6 Week4 0.1164 0.0657 0.2 0.8 Week1 0.0156 -0.4087 0.2 0.8 Sample SDe1 -0.695 0.2363 21.3 57.6 0.001 Depth SDe2 0.181 0.2716 4.6 13.2 SDe3 0.611 -0.5812 4.6 13.2

64 Appendix 2.1. Larval Lake Whitefish (Coregonus clupeaformis) Gape Size

Table A2.1. Gape size of larval Lake Whitefish (Coregonus clupeaformis) over eight weeks of development measured by horizontal measurement of the widest part of lower maxillary (limiting gape width) (See Figure A2.2). Where N= sample size, SD=standard deviation. Lower maxillary widths were measured using Image J software.

Week N Minimum Average SD Maximum (mm) (mm) (mm) 1 10 0.597 0.7321 0.0877 0.886 2 10 0.809 0.952 0.0986 1.076 3 10 0.823 1.0311 0.1406 1.288 4 10 0.851 1.0744 0.1067 1.232 5 10 1.018 1.2272 0.2461 1.907 6 10 1.035 1.2282 0.1036 1.373 7 10 1.211 1.4286 0.2112 1.805 8 5 1.276 1.4676 0.1455 1.675

65

Figure A2.1. Ventral view of larval Lake Whitefish (Coregonus clupeaformis) showing lower maxillary (limiting gape width) with line indicating widest part of gape. Scale is 1mm. Widths measured using Image J software.

66 Appendix 2.2. Larval Burbot (Lota lota).

Figure A2.2. Images of larval Burbot (Lota lota) collected on 05 July 2013 (Week 3) on the Douglas Point (DP) transect, station depth of 40m, at a sample depth of 15m in Lake Huron. Scale =1mm. (A – dorsal, B-ventral). Larval fish was identified as Lota lota by DNA Barcoding

67 Appendix 2.3. Environmental variables and particle size/zooplankton frequencies observed for all stations and samples over weeks 1-3 for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013.

68 A 20 15 10 5 1m 0

Temperature (°C ) 0 1 2 3 Week

B

20

15

10 1m

5 5m

Temperature (°C ) 0 0 1 2 3 Week C 20

15

10 1m 5 5m

Temperature (°C ) 0 15m 0 1 2 3 Week

D 20 15 10 1m 5 5m 0 0 1 2 3 15m Week Temperature (°C )

Figure A2.3.1 A-D: Temperature (oC) observed for all stations (3, 10, 20, 40m, A-D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3.

69 A

14 12

10 1m DO (mg/L) 8 0 1 2 3 Week

B 14

12

10 1m DO (mg/L) 5m 8 0 1 2 3 Week

C 14

12 1m 10 5m DO (mg/L) 8 15m 0 1 2 3 Week

D 14

12 1m 10 5m DO (mg/L) 8 15m 0 1 2 3 Week

Figure A2.3.2 A-D: Dissolved oxygen (mg L-1) observed for all stations (3, 10, 20, 40m, A-D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3.

70 A 3000 2500 2000 1500 1000 1m 500 0 Particle frequency 0 1 2 3 Week

B 3000 2000 1000 1m 0 5m 0 1 2 3

Particle frequency Week

C 3000

2000 1m 1000 5m 0 15m 0 1 2 3 Particle frequency Week

D 3000

2000 1m 1000 5m 0 15m

Particle frequency 0 1 2 3 Week

Figure A2.3.3 A-D: Total particle frequency observed for all stations (3, 10, 20, 40m, A- D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3.

71 A 1200 1000 800 600 400 3m Frequency Zoopankton 200 0 0 1 2 3 Week

B 1200 1000 800 600 400 1m Frequency Zooplankton 200 5m 0 0 1 2 3 Week

C 1200 1000 800 600 1m 400 5m Frequency

Zooplankton 200 0 15m 0 1 2 3 Week

D 1200 1000 800 600 1m 400 5m Frequency 200 Zooplankton 0 15m 0 1 2 3 Week

Figure A2.3.4 A-D: Total zooplankton frequency observed for all stations (3, 10, 20, 40m, A-D respectively) and samples (1, 5, 15m when available) for the Inverhuron (IH) transect at Douglas Point, Lake Huron in 2013 across weeks 1-3.

72 Appendix 2.4. Historic Sampling of phyto and nonicththyo-zooplankton at Douglas Point, Lake Huron.

Table A2.4. Plankton (phyto, zoo) species presence at Douglas Point, Lake Huron compiled from Johnson 1973; Wismer et al. 1986; Golder Associates 2008e. Sizes (µm), lengths and widths when available and morphological features for each genus/species provided when available. Species presenting extreme sexual dimorphism have been treated as two different species, one for male and one for female. * indicates most prominent genera present at the site.

Plankton Phylum Class/Order Genus/Species Maximum Size Morphology (µm)

Phyto Heterokonto Bacillariophyta /Pennales Tabellaria Length 39-83 -linear valves (diatoms) fenestrate* Width 2.1-6.9 -axial area is narrow and linear -septa are absent and marginal spines are absent. -The cells are joined in colonies that form long, straight chains. -frustules are rectangular in girdle view.

Asterionella spp.* Length 60-80 -star shaped Width 2-4 -8-20 celled colonies

73

Fragilaria spp.* Length 6-133.9 -rectangular Width 2-5.6 -joined by spines to form colonies

Stephanodiscus Length 13.8-120 -radial spp. -spines

Rhizosolenia spp. 2.5-170 -straight cells -slight curve -create large mats

Navicula spp. Length 12-42 -boat shaped Width 3-4

Nitzschia spp. Length 30-150 -keel present Width 3-8 -elongate

Melosira spp. Length 3-81 -tubular Width 5-47 -valve wall uniform -lacks spines

Synedra spp. Length 12-70 -linear Width 2.4-7.3 -needle like

Chrysophyta Dinobryon sp. 224 -vase like cells (golden algae)

74 -yellow-brown chloroplasts Cyanobacteria Cyanophyta/Chroococcales Chroococcus spp. 155 -2-8 cells (blue green algae) width 1.5-70 separated from each other -transparent sheath

Cyanophyta/Nostocales Anabaena spp. 4-50 -filamentous (blue green algae) - cylindrical/barrel shaped -looks like glass -produce neurptpxoms

Oscillatoria sp. Width 8-60 -filamentous -never branched -colonies

Aphanizomenon sp. Length 2cm -filamentous -looks like clumps of grass -pale blue-green -cells cylindrical

Chlorophyta Chlorophyceae -can be (green algae) unicellular -filamentous

75 Chlorophyceae/ Scenedesmus sp. Length 12.5 -colonial (4-8 Sphaeropleales Width 5 cell) -non-motile

Ankistrodesmus Width 0.8-6 -fiber shaped spp. Length 3-150 cells -needle like Chlorella sp. 2-10 -single cell -spherical Zoo Monogononta/Ploima Asplanchna spp.* 500-1500 -soft bodied Rotifera -sac like (rotifer) -very large -foot absent

Conochilus 290-840 -crown of cilia unicornis present -2 eyespots -feeding apparatus that contains jaw like 'trophi' Kellicottia 70-90 -long thin body longispina -long anterior (male) and posterior spines. -6 anterior spines of varying length.

Kellicottia 460-730 -long thin body longispina -long anterior

76 (female) and posterior spines. -6 anterior spines of varying length. Keratella sp.* Keratella Length 180 -footless rotifer cochlearis Width~50-60 -protective covering -tecta form lacks posterior spine -typica form has spine

Notholca foliacea 160-180 -tubular

Notholca squamula 152-196

Ploesoma 150-300 truncatum

Polyarthra sp.*

Polyarthra 80-160 -paddle like dolichoptera appendages Polyarthra 140 -4 sets of blade vulgaris projections -cilliated mouth -aloricate rotifer

77 Synchaeta oblonga 215-250 hard to detect lateral antennae, -eyes are unequal in size in some individuals -grasping rotifer -summer only

Synchaeta 200-250 -grasping rotifer pectinata 340-511 -year round

Synchaeta sp 275-300 -grasping rotifer

Arthropoda Branchipoda/Cladocera Bosmina 400-500 -laterally (Crustacea) longirostris* compressed (male) -spherical oblate spheroid body

Bosmina 400-600 -laterally longirostris* compressed (female) -spherical oblate spheroid body

Daphnia galeata 500-600 -shell spine mendotae -oval carapace -helmeted head -presence of rostrum

Eubosmina ~400 Postabdominal

78 coregoni claw with proximal pecten only, mucro absent, sensory bristle near tip of rostrum Holopedium 500-600 -humped gibberum carapace/brood (male) pouch -mantle 2x body

Holopedium 1500-2200 -humped gibberum carapace/brood (female) pouch -mantle 2x body

Polyphemus 800 short body, pediculus body and legs (male) not covered with bivalve carapace, four pairs of stout legs are flattened, clearly segmented and prehensile

Polyphemus 1500 short body, pediculus body and legs (female) not covered with bivalve

79 carapace, four pairs of stout legs are flattened, clearly segmented and prehensile

Maxillopoda/Calanoida Diaptomus minutes 900 -tapered head -single eye

Maxillopoda/Cyclopoida Diacyclops 800-1000 -short attenae bicuspidatus -paired thomasi mandibles (male) -caudal rami 3x long as wide -lateral seta -blunt head -single eye

Diacyclops 900-1600 -short attenae bicuspidatus -paired thomasi mandibles (female) -caudal rami 3x long as wide -lateral seta -blunt head -single eye

Tropocyclops 500-600 -fifth leg prasinus -3 terminal mexicanus spines

80 (male) -12 antennule segments -long caudal rami (3x long as broad)

Tropocyclops 500-900 -fifth leg prasinus -3 terminal mexicanus spines (female) -12 antennule segments -long caudal rami (3x long as broad)

81 Appendix 2.5. Larval Lake Whitefish (Coregonus clupeaformis) Diet

Table A2.5. Larval Lake Whitefish (Coregonus clupeaformis) diet by class/order and genus/species in field and laboratory with associated references. *mesocosm experiment **remained quantitatively the most important prey item Plankton Phylum Class/Order Genus/Species Field Laboratory Great Lakes Other

Zoo Arthropoda Maxillopoda/ Freeberg et al. 1990 Davis & Todd 1998* (Crustacea) Cyclopoida Johnson et al. 2009 Hoyle et al. 2011

Diacyclops thomasi Claramunt et al. 2010 Cyclops bicuspidatus Teska & Behmer 1981 Cyclops viridis Hart 1930a Branchipoda/ Johnson et al. 2009 Cladocera Hoyle et al. 2011 Bosmina spp.** Reckahn 1970 Davis & Todd 1998* Hoagman 1973 Claramunt et al. 2010 Bosmia longirostris Hart 1930a Davis & Todd 1998*

Chydorus sphaericus Davis & Todd 1998* Bythotrephes longimanus Macpherson et al. 2010 Daphnia spp. Hoagman 1973 Davis & Todd 1998 Pothoven & Nalepa 2006 Claramunt et al. 2010

82 Maxillopoda/ Freeberg et al. Davis & Todd 1998 Calanoida 1990*** Hoyle et al. 2011*** Leptodiaptomus sicilis Claramunt et al. 2010*** Diaptomus spp. Reckahn 1970 Raisanen & Behmer Claramunt et al. 1979 2010*** Diaptomus sicilis Teska & Behmer 1981 Diaptomus ashlandi Hart 1930a Limnocalanus macrurus Teska & Behmer 1981 Branchiopoda/ Artemia salina Fluchter 1980 Anostraca Taylor & Freeberg 1984 Drouin et al. 1986 Harris & Hulsman 1991 Bengtson et al. 1991 Harris 1992 Brown & Taylor 1992 Rellstab et al. 2004 Rotifera Davis & Todd 1998 (rotifer) (excluded less than 1%) Nothalca spp. Hoagman 1973 Frederick 1982*** Claramunt et al. 2010*** Keratella spp. Frederick 1982 Claramunt et al. 2010*** Benthic Insecta/Diptera Chironomidae Johnson et al. 2009 Invert

83 Chapter 3: Increased taxonomic resolutions of Laurentian Great Lakes ichthyoplankton through DNA barcoding: A case study comparison against visual identification of Stokes Bay, Lake Huron ichthyoplankton.

3.1. Abstract

Ichthyoplankton can be difficult to identify due to a lack of morphological differences among closely-related species during embryonic and larval development. Traditional visual identification of ichthyoplankton requires special knowledge to employ available morphological keys. Accurate species-level identification is required to support monitoring programs for those taxa that are considered valued ecosystem components in many North American freshwater habitats. DNA-based methods are gaining in popularity as an alternative to visual identification techniques, but direct comparisons of methodological performance are lacking. The goal of this study was to evaluate and explore the consistency between DNA barcoding and visual identification methods using a case study of larval fish caught in plankton tows at Stokes Bay, Lake Huron. Four participants served as ‘Identifiers’ (3 novices, 1 expert) for visual identification of this multi-species set of larval fishes; including Lake Whitefish, Lake Herring, Yellow Perch, Iowa Darter, Ghost Shiner, Rainbow Smelt. Visual identification performance varied from family to family, species to species and among observers. DNA barcoding successfully identified 51/55 (93%) of the larval fishes in a 2011 sample from Stokes Bay, Lake Huron. This study revealed there was a 94% consistency between visual identification and DNA barcoding at the family level with an average of 71.5% consistency between DNA barcoding and visual identification results to the species level. Results of this evaluation strongly support the use of DNA barcoding in combination with visual identification to improve the accuracy and precision of species identification for Great Lakes ichthyoplankton.

______

In review in Journal of Great Lakes Research and co-authored by E. Holm, S. S. Crawford and R. H. Hanner

84 3.2. Introduction

Fish assemblages are an extremely important indicator of aquatic ecosystem health

(Simon & Lyons 1995). Planktonic assemblages of larval fishes, ichthyoplankton, are typically composed of individuals from multiple species and are studied to assess habitat usage, spawning activity, and food web ecology in aquatic ecosystems (Boehlert &

Mundy 1993; Richardson et al. 2007). A ‘biological species’ can be considered as the set of individuals/populations that actually or potentially interbreed in the wild, leading to the production of offspring that in turn have the capacity to reproduce and are typically recognized based on morphology (Mayr 2000). Traditional ichthyoplankton studies require specialized knowledge of anatomy and ontogeny to employ dichotomous keys to facilitate specimen identification based on the visual inspection of morphological characteristics that are thought to reliably discriminate species. However, difficulties in level of taxonomic resolution arise when identifying larvae due to the overlap in body size distribution of different species and the potential ambiguity of meristic and morphometric characteristics used to tell them apart (Richardson et al. 2007). The use of ichthyoplankton sampling gear introduces additional complexities through the potential to cause physical damage to individual specimens, making it even more difficult to identify them (Ager et al. 2006). Because of these challenges, the results stemming from visual attempts to identify ichthyoplankton may include high error rates that could have major implications for the validity of early life history studies (Hare et al. 1994; Kochzius et al. 2008; Durand et al. 2010; Puncher et al. 2015), yet most studies assume (implicitly or explicitly) their visual identifications of species to be accurate. In this study, we test this assumption with a case study of Laurentian Great Lakes fishes caught in a larval

85 ecology study in Stokes Bay, Lake Huron for the first time using a more specific approach.

Molecular methods provide an alternative approach for specimen identification.

DNA barcoding – a method of species identification that utilizes sequence variation in the mitochondrial cytochrome c oxidase subunit 1 (COI) gene to discriminate species

(Hebert et al. 2003) has proven to be particularly useful for fishes (Ward et al. 2009) including those from North America (Hubert et al. 2008; April et al. 2012). Several authors have suggested that relative to other genes, COI is most successful at identifying closely related species and individuals to the species level (Steinke et al. 2009). Many stages in the barcoding workflow can also be automated (Hajibabaei et al. 2005;

Borisenko et al. 2009; Hebert et al. 2013), so the procedure can be scaled up to rapidly and efficiently analyze large numbers of specimens. Moreover, the Sanger sequencing technique employed as part of the barcoding workflow generally produces high quality, reliable sequence data with low error rates in the range of 0.001-1% bases misread

(Sanger et al. 2007; Hoff 2009), and it can be applied to degraded or damaged samples, because only a small amount of tissue is needed for the genetic analysis (Paine et al.

2007; Ward et al. 2009; Steinke et al. 2009).

Some authors, such as Hubert et al. (2008), have suggested that barcoding could greatly simplify identification of individual freshwater fishes during early life history.

Larval fish have been successfully identified using DNA barcoding in Australia (Pegg et al. 2006), among coral reefs in the Indo-Pacific (Hubert et al. 2010), along the Caribbean coast of Panama (Victor et al. 2009), and along the Yucatan Peninsula, Mexico (Valdez-

Moreno et al. 2010). However, few empirical studies have been undertaken to compare

86 the effectiveness and efficiency of visual and DNA barcoding identification techniques

(e.g., Ko et al. 2013; Becker et al. 2015; Puncher et al. 2015).

Ichthyoplankton in the subfamily Coregoninae are particularly difficult to identify due to a general lack of species-specific morphological differences during embryonic and larval periods (Scott & Crossman 1973; Auer 1982; Todd & Stedman 1989; Ward et al.

2009). These challenges in visual species identification are also true for some adult coregonines, when individuals have proven difficult to distinguish using traditional meristic and morphometric characteristics (Todd & Stedman 1989; Schlei et al. 2008;

Teletchea 2009). Coregonus clupeaformis (Lake Whitefish), C. artedi (Lake Herring), C. hoyi (Bloater) and Prosopium cylindraceum (Round Whitefish) are important species commercially and ecologically in the Great Lakes (Roth et al. 2012). Both C. clupeaformis and P. cylindraceum have been identified and classified as valued ecosystem components in Lake Huron because of their ecological significance, economic value and having important spawning locations in Lake Huron (Holmes & Noakes 2002;

Brown 2007; Brown 2008); therefore, their accurate identification is important for Great

Lakes fisheries management. C. clupeaformis, which is a significant species to both

Aboriginals and to commercial fisheries (Crawford et al. 2001), will be the focus of this study. Because Lake Whitefish feature in ongoing monitoring programs, they provide a useful test case.

The goal of this study was to evaluate the precision and accuracy of species-level identification among multiple visual identifiers and compared with DNA barcoding. A subset of Laurentian Great Lakes ichthyoplankton was used as a case study of ichthyoplankton collected at Stokes Bay, Lake Huron. In order to achieve this goal, the following objectives were satisfied. A collection of ichthyoplankton was undertaken

87 during an ecological sampling regime of Laurentian Great Lakes ichthyoplankton. The results of visual species-level identification for each individual in a case study sample of ichthyoplankton, by novice/expert analysts using a microscope and an established morphological key was compiled. For each individual in the ichthyoplankton sample, the

COI gene was sequenced, and these sequences were cross-referenced against COI sequences previously published. Effectiveness/efficiency of visual identification, relative to DNA barcoding identification for the ichthyoplankton sample were contrasted.

3.3. Materials and Methods

Ichthyoplankton were sampled at Stokes Bay, Lake Huron from 21 April – 08

July 2011. Sampling began as the ice retreated and the bay was accessible and safe to navigate by boat. Ichthyoplankton were collected using a 500-micron nylon mesh plankton net equipped with a dolphin adaptor and bucket (0.75m hoop diameter, 2.3m net length). Each tow ran for five minutes at a speed ranging between 1.6-2.5 kph and covered an average distance of 360 m. Directionality was based on wind and wave conditions at each site; tows were directed towards prevailing wind/wave direction.

Larvae collected from the tows were stored in 95% ethanol. Five fish were randomly selected from each week of the sampling season for a total of n=55 larvae, each of which was transferred to a separate, numbered vial. Voucher specimens were deposited at the

Royal Ontario Museum, except for specimens 19 and 47, which were lost during identification (Appendix 3.1).

Four anonymous participants (hereafter referred to as ‘Identifiers’), attempted to visually identify the specimens to species level following an established dichotomous key for identification of Great Lakes larval fishes (Auer 1982), based on mensural and

88 meristic morphological characteristics (Figure 3.1). Identifiers #1-3 were deemed novices, with some previous experience (<5 years) in fish identification (e.g., training workshops with adult fishes, identification of ichthyoplankton from previous sampling programs), while Identifier #4 was deemed an expert (over 30 years) in taxonomy and identification of adult North American freshwater fishes. Identifiers 1-3 used an Olympus

SZ40-CTV binocular microscope and Infinity 2 microscope camera; Identifier 4 used a

Wild-Leitz binocular microscope. Visual identifiers could either identify specimen to species, to family, unknown (attempted to identify and could not make a confident decision) or not attempted (fish either missing from sample during time of identification, or not visually identified).

For DNA barcoding, caudal fin clips were taken from each specimen (after examination by visual Identifiers) for DNA extraction using Qiagen DNeasy Blood and

Tissue Kits (Qiagen, Valencia, CA) following the manufacturers instructions. PCR amplification was carried out in 12.5 µl reaction volumes with 6.25 µl 10% trehalose, 2

µl ddH2O, 1.25 µl 10X buffer, 0.625 µl MgCl2 (50 mM), 0.1 µl each of forward and reverse 10 µM concentration PCR primers (C_VF1LFt1 and C_VR1LRt1; Ivanova et al.

2007), 0.0625 µl 10 mM dNTPs, 0.06 µl Platinum Taq polymerase (5 U/µl), and 2 µl of

DNA. The reaction profile was an initial hot start at 94°C for 120 secs; followed by five cycles with a denaturation at 94°C for 30 sec, annealing temperature of 50°C for 40 sec, and extension at 72°C for 1 min; followed by thirty five cycles with a denaturation at

94°C for 30 seconds, annealing temperature of 54°C for 40 seconds, and extension at

72°C for 1 minute; and a final extension at 72°C for 10 minutes and hold at 4°C indefinitely. Amplification success of PCR products was ascertained visually using pre-

89 cast 2% Agarose E-gels (Invitrogen), where a clear, unambiguous band on the gel indicated successful amplification.

DNA sequencing reactions were carried out in 14 µl reaction volumes with 1 µl

BigDye v3.1, 1 µl 5X SeqBuffer, 1 µl primer (10 µM; C_VF1LFt1 for forward and

C_VR1LRt1 for reverse), 10 µl ddH2O, and 1.5 µl PCR product. PCR products were bidirectionally sequenced using an ABI 3730 DNA Analyzer (Applied Biosystems). Bi- directional sequence contig assemblies were created and edited using Sequencher v. 4.9

(Gene Codes), and multiple sequence alignments were generated manually using BioEdit v. 7.0.5.3 (Hall 1999). Sequences were uploaded to the Barcode of Life Data System

(BOLD; Ratnasingham & Hebert 2007) public project titled “Stokes Bay, Ontario, Lake

Whitefish” (project code: SBOLW) and subsequently submitted to GenBank (accession numbers KP978018-KP978067) via BOLD. All barcodes were queried against the sequences of known provenance using the BOLD ID engine (www.boldsystems.org,

Ward et al. 2009) using the “Species ID” option. A DNA barcode species match was defined as a percent of sequence similarity greater than 99% to a unique species in the reference database. Consistency in species identification between DNA barcoding and the group of visual Identifiers was conducted using a chi-square test coded in R (R version

2.13.1, 2011-07-08).

3.4. Results

Of the 55 fish larvae subsampled from the 2011 Stokes Bay ichthyoplankton collection, 51 (92.7%) were successfully amplified and sequenced by DNA barcoding and identified in BOLD (Table 3.1). Low quality forward and reverse sequences were obtained for one specimen and discarded, while three other specimens did not yield

90 sequences. Of the remaining 51 specimens identified by DNA barcoding in this study, 15

(29.4%) were identified as Coregonus artedi (Lake Herring), 27 (52.9%) were identified as C. clupeaformis (Lake Whitefish), 6 (11.8%) were identified as Etheostoma exile

(Iowa Darter) and 1 (1.9%) each were identified as Perca flavescens (Yellow Perch),

Osmerus mordax (Rainbow Smelt) and Notropis buchanani (Ghost Shiner).

Variation among visual identifiers was evident. Of the 55 larvae visually identified, precision between the four Identifiers was 45% (25/55), with concordant identifications. Compared to specimens that yielded high quality sequences, consistency among Identifiers were not consistent at 30/51 (59%), with one or more Identifier identifying a larval specimen differently from other Identifiers. Identifiers 1 and 2 processed all specimens in the sample, with 62.7% (32/51) and 72.5% (37/51) consistency with barcoding, respectively. Identifier 3 exhibited 83.3% (35/42) consistency with DNA barcoding species-level identification; however the identifier declined to identify 9 specimens, all of which were identified by DNA barcoding as non- coregonine species. Identifier 4 exhibited 91.7% (33/36) consistency with DNA barcoding species-level identification; however, identifier declined to identify a different set of 9 specimens to species, but instead provided a family-level identification for 7 of these specimens.

Overall, at the family level there was a 93.7% (181/193) consistency between visual and DNA barcoding identification (Table 3.1, see final column). There was a single identifier inconsistency for one of the 42 specimens identified by DNA barcoding as belonging to Salmonidae. For the seven Percidae, two of the three novice identifiers exhibited complete consistency with DNA barcoding, while the remaining novice identifier was not able to identify to family. The expert identifier did not identify four of

91 the percids to family level; of the remaining three specimens, one was consistent with

DNA barcoding, one was inconsistent, and one was not attempted to be identified. For the single specimens identified by DNA barcoding as Osmeridae and Cyprinidae, there was zero consistency among the visual identifiers.

At the species level, there was a 71.5% (138/193) consistency between visual and

DNA barcoding identification. For the 15 specimens identified by DNA barcoding as

Lake Herring, there was 63.3% (38/60) consistency with visual identification. For the 27 specimens identified by DNA barcoding as Lake Whitefish, there was a 90.7% (98/108) level of consistency with visual identification. For the 6 specimens identified by DNA barcoding as Iowa Darter, there was zero consistency since all visual identifications were recorded as Yellow Perch. The single specimen identified by DNA barcoding as Yellow

Perch was consistently identified by both of the novice visual Identifiers who provided a species-level designation. Chi-square tests indicated that the group of visual Identifiers were not significantly different among themselves in species-level consistencies (p=

0.1351, df=3); however, the difference between DNA barcoding and visual identifications was significant (p=0.004, df=4).

3.5. Discussion

The goal of this study was to evaluate the consistency of visual identifications of

Laurentian Great Lakes ichthyoplankton specimens among a set of four researchers as compared to results obtained from DNA barcoding. Overall, the 71.5% consistency observed between the two identification methods was relatively high, considering that (a) three of the four visual Identifiers had limited experience with larval fish identification, and (b) the Identifiers were working with a first edition of a dichotomous identification

92 key for Great Lakes larval fishes (Auer 1982). However, even with a reasonably high consistency between DNA barcoding and visual identification, it is important to keep in mind a variety of factors when considering current and future applications of these techniques for identifying species of ichthyoplankton.

Given the definition of a ‘biological species’ adopted in this study, we must acknowledge the ever-present potential for inter-specific hybrids in samples from wild ecosystems. DNA barcoding and most visual identification keys operate under the assumption (implicit or explicit) that the specimen in question is not the product of hybridization. However, we know from experience that hybridization can occur at non- trivial levels in the wild (Scribner et al. 2001; Keck & Near 2009), especially among closely related species-flocks such as the coregonines (Turgeon et al. 1999). Both DNA barcoding and visual identification methods are vulnerable to mis-interpreting naturally occurring hybrids. Barcoding cannot effectively identify hybrids due to the maternal inheritance of the mitochondrial genome (Strugnell 2007) although it is possible that sequencing a nuclear gene could help improve accuracy and precision of hybrid identifications. Visual identification of hybrids is usually only possible if there is sufficient distinction between parental species in the states of morphological characteristics used to discriminate them (Keck & Near 2009). From an operational perspective, overlapping distributions of morphological states tend to increase uncertainty for decision-making at nodes in a dichotomous key, leading in turn to increased risk of error in species/hybrid identification. Specifically with regard to Lake Herring and

Bloater in the subfamily Coregoninae, Todd and Stedman (1989) concluded that hybrids can be very difficult to discriminate on the basis of morphological characteristics that typically serve to distinguish the parental species. Hubbs (1955) suggested that

93 hybridization between Lake Whitefish and Lake Herring rarely occurs in nature, but

Garside and Christie (1962) did report low frequencies of this hybrid under artificial conditions, and seven individuals of this hybrid have been recently identified in inland lakes of Ontario (ROM, unpublished data). For other species identified by DNA barcoding in this study, Keck and Near (2009) reported that hybridization of darters

(Percidae: Etheostomatinae) is rare, while Wiggins et al. (1983) observed that Yellow

Perch x Sander vitreus (Walleye) hybrid embryos failed to hatch in a laboratory environment. Taken together, these reports strongly suggest that while it is indeed possible that hybrids existed in our sample from Lake Huron, there is a low probability that any observed inconsistencies between DNA barcoding and visual identification would be explained by this factor.

While DNA barcoding for species-level identification relies on the assumption that DNA barcoding has high fidelity with biological species (Hebert et al. 2003; Song et al. 2008; Puncher et al. 2015), there are actually several possible sources of error that can occur during the processing phases of DNA barcoding. Error can occur in the preliminary stages of specimen handling (e.g. specimen storage, preservation, DNA degradation), in the pre-processing stage (e.g. tissue collection, contamination, reagent pipetting) and in the post-processing stage (e.g. sequence editing and alignment, sequence upload and matching on BOLD). In our study, all protocols and procedures were followed closely to minimize sources of processing error during DNA barcoding. Out of

55 larval fishes on which barcoding was attempted, 51 successfully yielded high quality sequences that provided a species-level match on BOLD (Table 3.1). One low quality sequence was removed from further consideration, while three samples failed to amplify.

Other genetics methods (i.e. mini-barcodes; Hajibabaei et al. 2006) could have been used

94 to recover DNA from these failures, we chose to report our results following a standard barcoding protocol widely applied to North American freshwater fishes (Hubert et al.

2008). It is important to note that the single specimen identified as a Ghost Shiner could in fact be a Mimic Shiner (Notropis volucellus); previous efforts were unable to detect differences between these two species due to shared barcode haplotypes (Hubert et al.

2008). Mimic Shiner is more likely in Stokes Bay, although Ghost Shiner may be present but would be a range extension based on the distribution map in Holm et al. (2009, 2010).

Visual identification methods such as the Auer (1982) dichotomous key, rely on the assumption that the target species can be effectively discriminated on the basis of (a) non overlapping distribution of a single morphological characteristic, and/or (b) clear support of one set of morphological characteristics over an alternate set. In this preliminary case study, there was little evidence to suggest that Identifiers of different skill level differed substantially in their ability to use the dichotomous key to identify ichthyoplankton to species. However, it is always possible that Identifiers could have mis-interpreted the intended morphological states used in the decision nodes and/or mis- perceived the actual morphological states of the specimen. Table 3.2 presents the actual sequence of morphological decision nodes used in the Auer (1982) dichotomous key to discriminate among the species identified by DNA barcoding in this study. If DNA barcoding identification in this study was assumed to be correct, then a mistake in visual species identification could have occurred at any one of the 12 different nodes leading to

Family Salmonidae (i.e. nodes 1/2/3/5/6/7/8/9/10/21/22/23). Obviously, the number and operational difficulty of decision nodes is a very important factor in assessing the risk

(probability and magnitude) of mis-identification error when using any dichotomous key

95 based on morphological characteristics. We see this as a powerful approach for evaluating the utility of dichotomous keys more broadly.

In the future, we recommend that both DNA barcoding and visual identification continue to be used together for identification of ichthyoplankton, since using a combination of methods can safeguard against method-specific identification biases and errors (Puncher et al. 2015). We also strongly recommend that analysts take advantage of the fact that we are rarely completely ignorant about the species and life-histories of specimens in a given ecological context. Prior probabilities for species occurrence at a given sampling location could be established on the basis of historical samples and general species-specific patterns in habitat, spawning, and early-life history (e.g. Balon

1975, 1987). One should consider the cost effectiveness, materials, labour and analysis for the desired method. For this study, we estimated that the DNA barcoding would have required approximately 35 hours for all samples at a cost of $10 a specimen while visual identification by a single analyst required approximately 20 hours at an estimated cost of

$50-100 per specimen (Greenaway, pers. comm. Golder Associates. 2012). Clearly it would be important for these kind of metrics to be evaluated on a case by case basis.

Finally, we believe that genetic and visual identification methods should continue to be deployed concurrently in anticipation of future changes in distribution and abundance of species in the wild - whether these changes result from natural or anthropogenic causes.

Concluding Remarks

Both DNA barcoding and visual identification are time-consuming identification methods for individual fish; however, DNA barcoding had a high recovery rate (92.5%)

96 relative to visual identification, where some visual identifiers declined identification of samples. Visual identification on average had a 74.0% success rate for Lake Whitefish and Lake Herring, and an average of 10% success rate for other species, and an overall consistency of 71.5% when compared to DNA barcoding for species-level identifications.

By increasing the accuracy of ichthyoplankton identification, environmental assessments and monitoring will be improved (Teletchea 2009). Studies of recruitment success, ecology of fishes, habitat selection of marine and freshwater fishes and entrainment estimates of water usage industries will also be improved. This seems likely given the trend to high throughput and low cost observed since the inception of DNA barcoding.

97 3.6. References

Ager D.D., Cord I., Patric, P.H. 2006. Entrainment sampling at Darlington Nuclear Generating Station – 2004. Report Number NK38-REP-07264-10001-R00.

April J., Mayden R.L., Hanner R.H., Bernatchez L. 2012. Genetic calibration of species diversity among North America’s freshwater fishes. PNAS 108(26): 10602- 10607.

Auer N.A. (ed.) 1982. Identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage. Great Lakes Fishery Commission, Ann Arbor, MI 48105. Special Pub. 82-3:744 pp.

Balon E.K. 1975. Reproductive guilds of fishes: a proposal and definition. Journal of Fish Research Board Canada 32: 821-864.

Balon E.K. 1987. On decisive events in the early life history of fishes and the pertinence of such events to the functional grouping of fishes according to their susceptibility to impact. Section 5. 18pp.

Becker R.A., Sales N.G., Santos G.M., Santos G.B., Carvalho D.C. 2015. DNA barcoding and morphological identification of neotropical ichthyoplankton from the Upper Parana and Sao Francisco. Journal of Fish Biology doi:10.1111/jfb.12707.

Boehlert G.W., Mundy B.C. 1993. Ichthyoplankton assemblages at seamounts and oceanic islands. Bulletin of Marine Science 53 (2): 336-361(26).

Borisenko A.V., Sones J.A., Hebert P.D.N. 2009. The front-end logistics of DNA barcoding: challenges and prospects. Molecular Ecology Resources 9 (Suppl. 1): 27-34.

Brown T. 2007. Whitefish Investigations 2006 Summary B-REP-00531-00017. 7pp.

Brown T. 2008. Whitefish Investigations 2007 Summary B-REP-00531-00022. 5pp.

Crawford S., Muir A., McCann K. 2001. Ecological basis for recommendation of 2011 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Report prepared for the Chippewas of Nawash First Nation. 168pp.

Durand J.D., Diatta M.A., Diop K. Trape S. 2010. Multiplex 16S rRNA haplotype- specific PCR, a rapid and convenient method for fish species identification: an application to West African Clupeaform larvae. Molecular Ecology Resources 10: 568- 572.

Garside E.T., Christie W.J. 1962. Experimental hybridization among three coregonine fishes. Transactions of the American Fisheries Society 91(2): 196- 200.

98 Hajibabaei M., deWaard J.R., Ivanova N.V., Ratnasingham S., Dooh R.T., Kirk S.L., Mackie P.M., Hebert P.D.N. 2005. Criticial factors for assembling a high volume of DNA barcodes. Philosophical transactions of the Royal Society 360: 1959-1967.

Hajibabaei M., Smith M. A., Janzen D. H., Rodriguez J. J., Whitfield J. B., Hebert P. D. N. 2006. A minimalist barcode can identify a specimen whose DNA is degraded. Molecular Ecology Notes 6: 959-964.

Hall T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41:95–98 .

Hare J.A., Cowen R.K. Zehr J.P., Juanes F., Day K.H. 1994. Biological and oceanographic insights from larval labrid (Pisces: Labridae) identification using mtDNA sequences. Marine Biology 118: 17-24.

Hebert P.D.N., Cywinska A., Ball S.L., deWaard J.R. 2003. Biological identifications through DNA barcodes. Proceedings: Biological Sciences 270(1512): 313-321.

Hebert P.D.N., deWaard J.R., Zakharov E.V., Prosser S.W.J., Sones J.E., McKeown J.T.A., Mantle B., La Salle J. 2013. A DNA ‘Barode Blitz’: Rapid Digitization and Sequencing of a Natural History Collection. PLoS One 8(7): e68535.

Hoff K.J. 2009. The effects of sequencing errors on metagenomic gene prediction. BMC Genomics 10:520.

Holm E, Mandrak N.E., Burridge M.E. 2009, 2010. A ROM Field Guide to the Freshwater Fishes of Ontario, Royal Ontario Museum.

Holmes J.A., Noakes D. L. G. 2002. Whitefish Interactions with Nuclear Generating Stations: Recommendations for Monitoring the Effects of the Bruce Nuclear Power Development on Whitefish in Lake Huron. 173 pp.

Hubbs C.L. 1955. Hybridization between fish species in nature. Systematic Zoology 4(1): 1-20.

Hubert N., Hanner R., Holm E., Mandrak N.E., Taylor E., Burridge M., Watkinson D., Dumont P., Curry A., Bentzen P., Zhang J. April J., Bernatchez L. 2008. Identifying Canadian freshwater fishes through DNA barcodes. PlosOne 3(6):1-8.

Hubert N., Delrieu-Trottin ,E., Irisson , J-O., Meye, C., Planes, S. 2010. Identifying coral reef fish larvae through DNA barcoding: A test case with the families Acanthuridae and Holocentridae. Molecular Phylogenetics and Evolution 55: 1195-1203.

Ivanova N.V., Zemlak T. S., Hanner R. H., Hebert P.D.N. 2007. Universal primer cocktails for fish DNA barcoding. Molecular Ecology Notes 7: 544-548.

99 Keck B.P., Near T.J. 2009. Patterns of Natural Hybridization in Darters (Percidae: Etheostomatinae). Copeia (4): 758-773.

Ko H-L, Wang Y-T., Chiu T-S., Lee M-A., Leu M-Y., Chang K-Z., Chen W-Y., Shao K- T. 2013. Evaluating the accuracy of morphological identification of larval fishes by applying DNA barcoding. PloS One 8(1): e53451.

Kochzius M., Nolte M., Weber H., Silkenbeumer N., Hjorleifsdottir S., Hreggvidsson G.O., Marteinsson V., Kappel K., Planes S., Tinti F., Magoulas A., Garcia-Vazquez E., Turan C., Hervet C., Camp Falgueras D., Antoniou A., Landi M., Blohm D. 2008. DNA microarrays for identifying fishes. Mar. Biotechnol. 10: 207-217.

Mayr E. 2000. The Biological Species Concept in Sepceis Concepts and Phylogentic Theory: A Debate. Wheeler, Q. D and Meier, R. (eds). Columbia University Press.

Paine M.A., McDowell J.R., Graves J.E. 2007. Specific identification of Western Atlantic ocean scombrids using mitochondrial DNA cytochrome c oxidase subunit 1 (COI) gene region sequences. Bulletin of Marine Science 80(2): 353-367.

Pegg G.G., Sinclair B., Briskley L., Aspden W.J. 2006. MtDNA barcode identification of fish larvae in the southern Great Barrier Reef-Australia. Scientia Marina 70(S2): 7-12.

Puncher G. N., Alemany F., Arrizabalaga H., Cariani A., Tinti F. 2015. Misidentication of Bluefin tuna larvae: a call for caution and taxonomic reform. Reviews in Fish Biology and Fisheries. DOI 10.1007/s11160-015-9390-1.

R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing,Vienna, Austria. ISBN 3-900051-07- 0, URL http://www.R-project.org/.

Ratnasingham S., Hebert P.D.N. 2007. Barcoding BOLD: The Barcode of Life Data System (www.barcodinglife.org). Molecular Ecology Notes 7: 355-364.

Richardson D.E., Vanwye J.D., Exum A.M., Cowen R.K., Crawford D.L. 2007. High throughput species identification: from DNA isolation to bioinformatics. Molecular Ecology Notes 7: 199-207.

Roth B.M., et al. 2012. Fishes and Decapod Crustaceans of the Great Lakes Basin. In: Great Lakes Policy and Management: A Binational Perspective. Eds. Taylor, W.W. and A. Lynch. Michigan State University Press.

Sanger F., Nicklen S., Coulson A. R. 1977. DNA sequencing with chain-terminating inhibitors. PNAS 74(12): 5463-5467.

Schlei O.L., Crete-Lafreniere A., Whiteley A.R., Brown R.J., Olsen J.B., Bernatchez L.,Wenburg J.K. 2008. DNA barcoding of eight North American coregonine species. Molecular Ecology Resources 8: 1212-1218.

100 Scott W.B., Crossman E.J. 1973. Freshwater Fishes of Canada. Fisheries Research Board of Canada, Ottawa. 966 pp.

Scribner K.T., Page K.S., Barton M.L. 2001. Hybridization in freshwater fishes: A review of case studies and ctoniclear methods of biological inference. Reviews In Fish Biology and Fisheries 10: 293-323.

Simon T.P., Lyons J. 1995. Application of the index of biotic integrity to evaluate water resource integrity in freshwater ecosystems. In: Davis, W.S., Simon, T.P. (eds.) Biological assessment and criteria: tools for water resource planning and decision- making. Boca Raton: Lewis Publishers, pp. 245-262.

Song H., Buhay J.E., Whiting M.F., Crandall K.A. 2008. Many species in one: DNA barcoding overestimates the number of species when nuclear mitochondrial Pseudogenes are coamplified. PNAS 105 (36): 13486-13491.

Steinke D., Zemlak T.S., Boutillier J.A., Hebert, P.D.N. 2009. DNA barcoding of Pacific Canada’s fishes. Marine Biology 156: 2641-2647.

Strugnell J.M., Lindgren A.R. 2007. A barcode of life database for the Cephalopoda? Considerations and concerns. Reviews in Fish Biology and Fisheries. 17: 337-344.

Teletchea F. 2009. Molecular identification methods of fish species: reassessment and possible applications. Reviews in Fish Biology and Fisheries 19: 265-293.

Todd T.N., Stedman R.M. 1989. Hybridization of ciscoes (Coregonus spp.) in Lake Huron. Canadian Journal of Zoology 67: 1679-1685.

Turgeon J., Estoup A., Bernatchez L. 1999. Species Flock in the North American Great Lakes: Molecular Ecology of Lake Nipigon Ciscoes (Teleostei: Coregonidae: Coregonus). Evolution 53: 1857-1871.

Valdez-Moreno M., Vasquez-Yeomans L., Elias-Gutierrez M., Ivanova N.V., Hebert P.D.N. 2010. Using DNA barcodes to connect adults and early life stages of marine fishes from the Yucatan Peninsula, Mexico: potential in fisheries management. Marine and Freshwater Research 61(6): 655-671.

Victor B. C., Hanner R., Shivji M., Hyde J., Caldow C. 2009. Identification of the larval and juvenile stages of the Cubera Snapper, Lutjanus cyanopterus, using DNA barcoding. Zootaxa 2215: 24-36.

Ward R.D., Hanner R., Hebert P.D.N. 2009. The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology 74: 329-356.

Wiggins T.A., Bender Jr. T.R., Mudrak V.A., Takacs M.A. 1983. Hybridization of Yellow Perch and Walleye. The Progressive Fish-Culturist 45(2): 131-132.

101

Figure 3.1. Diagram of a larval Lake Whitefish (Coregonus clupeaformis) indicating morphological characteristics, especially mensural and meristic features used in a dichotomous key for ichthyoplankton of the Great Lakes basin (Adapted from Faber 2006-13 www.fishbabies.ca).

102 Table 3.1. Consistency of species-level visual identification compared to DNA barcoding of ichthyoplankton. N= novice, E= expert. UN=unknown, FM=family, na=not attempted. Dark grey squares are incorrect identifications, light grey squares are not attempted, medium grey squares are identified correctly to family. Letters in boxes are what the identifier originally identified the specimen as.

DNA Barcoding Identification Specimen Code Visual Identification Consistency Family Species N1 N2 N3 E4 NO BARCODE Subthreshold (CC) 1 CC CC CC CC 4/4 NO BARCODE No sequence 2 CA CA CA CA na NO BARCODE No sequence 3 CA CC CA FM na NO BARCODE No sequence 4 PF PF na FM na Salmonidae Coregonus artedi = CA 1 CA CA CA CA 4/4 2 CA CA CA CA 4/4 3 CC CA CA CA 3/4 4 CC CA CA CA 3/4 5 CC CA CA CA 3/4 6 CA CA CA FM 3/4 7 CA CA CA UN 3/4 8 CA CC CA CA 3/4 9 CA CC CA CA 3/4 10 CA CA CC CA 3/4 11 CA CA CC CA 3/4 12 CC CA CC CA 2/4 13 CA CC CC CC 1/4 14 CC CC CC CC 0/4 15 CC CC CC CC 0/4 Coregonus clupeaformis = CC 16 CC CC CC CC 4/4 17 CC CC CC CC 4/4 18 CC CC CC CC 4/4 19 CC CC CC CC 4/4

103 20 CC CC CC CC 4/4 21 CC CC CC CC 4/4 22 CC CC CC CC 4/4 23 CC CC CC CC 4/4 24 CC CC CC CC 4/4 25 CC CC CC CC 4/4 26 CC CC CC CC 4/4 27 CC CC CC CC 4/4 28 CC CC CC CC 4/4 29 CC CC CC CC 4/4 30 CC CC CC CC 4/4 31 CC CC CC CC 4/4 32 CC CC CC CC 4/4 33 CC CC CC CC 4/4 34 CC CC CC CC 4/4 35 CC CC CC CC 4/4 36 CC CC CC na 3/3 37 CA CC CC CC 3/4 38 CA CC CC CC 3/4 39 CA CC CC CC 3/4 40 CA CC CC CC 3/4 41 CC CC CA FM 2/4 42 CA CA CC FM 1/4 Percidae Etheostoma exile = EE 43 PF PF na UN 0/3 44 PF PF na UN 0/3 45 PF PF na FM 0/3 46 PF PF na FM 0/3 47 PF PF na UN 0/3 48 PF PF na na 0/2 Perca flavescens = PF 49 PF PF na UN 2/3

104 Osmeridae Osmerus mordax = OM 50 PF PF na UN 0/3 Cyprinidae Notropis buchanani = NB 51 PF PF na UN 0/3 Total 138/193

105 Table 3.2. Frequency and use of different morphological characteristics in Auer's (1982) dichotomous key for ichthyoplankton of the Great Lakes basin, used by novice/expert identifiers in this study to determine taxonomic identification of ichthyoplankton collected from Stokes Bay, Ontario, Lake Huron (species identification for larvae of Subfamily Coregoninae). Grey shading indicates Family/Subfamily/species in the sample, as determined by DNA barcoding.

Key Morphological Characteristic Sample Node Variable Type Consequence

Provisional key to Families of Great Lakes larval fishes (p.15) 1 body shape quality 2 gill opening # (->2) meristic Other 2 chin barbels occurrence 3 Other 3 adhesive disc occurrence 5 Other 5 preanal length mensural 6 preanal myomere # meristic Other 6 pre/post-anal myomere # meristic/meristic 7 Other 7 pre/post-anal myomere # meristic/meristic 8 Other 8 postanal myomere # meristic 9 Other oil globule # meristic 9 mouth location quality 10 eye/total length diameter mensural/mensural Other gut termination location quality pectoral bud size quality 10 mouth location quality 21 swim bladder occurrence Other pigment location quality total myomere # meristic 12 preanal/total length mensural/mensural 13 preanal myomere # meristic Other 13 postanal myomere # meristic 16 Other 16 total myomere # meristic 18 Other 18 total myomere # meristic 20 preanal myomere # meristic Other oil globule occurrence CYPRINIDAE 20 yolk sac shape quality PERCIDAE anus location quality 21 total myomere # meristic 22

106 Other 22 preanal myomere # meristic 23 Other total length mensural 23 yolk sac diameter/total length mensural/mensural SALMONIDAE yolk sac pigmentation 24 dorsal pigmentation quality quality postanal myomere # meristic 24 body shape quality OSMERIDAE yolk sac size-shape-location quality Other total length mensural

Family SALMONIDAE yolk-sac larvae less than 20 mm total length (p.80) yolk sac/total length mensural/mensural 1 eye diameter/total length mensural/mensural COREGONINAE pigment density/location quality SALMONINAE body depth at anus/total mensural/mensural length

Subfamily COREGONINAE from hatching to first fin ray development (p.81) total length mensural oil globule # mersitic 1 oil globule diameter/yolk sac mensural/mensural P. cylindraceum length 2 melanophore size-shape- quality location-distribution myomere width mensural 2 total length mensual preanal myomere # meristic C. clupeaformis melanophore shape-location- quality 3 distribution melanophone/myomere mensural/mensural width 3 melanophore/myomere width mensural/mensural C. artedi C. hoyi

Subfamily COREGONINAE (from first fin ray development to stage when all median fins basally rayed, pelvic buds appear toward end of stage) (p.81) 1 total length mensural 2 3 2 melanophore #-shape- quality P. cylindraceum location C. clupeaformis melanophore/myomere width mensural/mensural snout-dorsal fin/standard

107 length mensural/mensural 3. melanophore shape-size quality C. artedi melanophore/myomere width mensural/mensural C. hoyi

Subfamily COREGONINAE (full complement of median fin rays appear and pelvic buds or fins present) (p.82) 1 parr marks quality P. cylindraceum melanophore clusters quality 2 total length mensural 2 lower jaw/snout length mensural/mensural predorsal/standard length mensural/mensural C. clupeaformis preanal myomere # meristic C. artedi/hoyi total length mensural

occurrence = 4 Frequency of Morphological quality = 19 Characteristic Types meristic = 20 mensural = 37

108 3.7. Appendices

Appendix 3.1.

Table A3.1. Royal Ontario Museum Catalogue Numbers of Voucher Specimens

Number BOLD ID vial TCN BOX POS 1 Coregonus clupeaformis 1A T16371 I221 B3 2 Coregonus clupeaformis 1B T16372 I221 B4 3 Coregonus clupeaformis 1C T16373 I221 B5 4 Coregonus clupeaformis 1D T16374 I221 B6 5 Coregonus clupeaformis 1E T16375 I221 B7 6 Coregonus artedi 1F T16377 I221 B9 7 Coregonus artedi 1G T16378 I221 C1 8 Coregonus artedi 1H T16379 I221 C2 9 Coregonus clupeaformis 1I T16376 I221 B8 10 Coregonus clupeaformis 2A T16384 I221 C7 11 Coregonus clupeaformis 2B lost - - 12 Coregonus clupeaformis 2C T16386 I221 C9 13 Coregonus clupeaformis 2D T16387 I221 D1 14 Coregonus clupeaformis 2E T16388 I221 D2 15 no barcode 2F T16463 I221 H2 16 Coregonus artedi 2G T16380 I221 C3 17 Coregonus artedi 2H T16381 I221 C4 18 Coregonus artedi 2I T16382 I221 C5 19 Coregonus artedi 2J T16383 I221 C6 20 Coregonus clupeaformis 3A lost - - 21 Coregonus clupeaformis 3B T16390 I221 D4 22 Coregonus clupeaformis 3C T16391 I221 D5 23 Coregonus clupeaformis 3D T16435 I221 E1 24 Coregonus clupeaformis 3E T16436 I221 E2 25 Coregonus clupeaformis 3F T16393 I221 D7 26 Coregonus clupeaformis 3G T16392 I221 D6 27 Coregonus artedi 3H T16437 I221 E3 28 Coregonus clupeaformis 3I T16433 I221 D8 29 Coregonus clupeaformis 3J T16434 I221 D9 30 Coregonus clupeaformis 4A T16438 I221 E4 31 Coregonus clupeaformis 4B T16439 I221 E5 32 Coregonus artedi 4C T16440 I221 E6 33 Coregonus clupeaformis 4D T16442 I221 E8 34 Coregonus clupeaformis 4E T16443 I221 E9 35 Coregonus artedi 4F T16447 I221 F4 36 Coregonus artedi 4G T16441 I221 E7 37 Coregonus clupeaformis 4H T16444 I221 F1 38 Coregonus clupeaformis 4I T16445 I221 F2

109 39 Coregonus clupeaformis 4J T16446 I221 F3 40 Coregonus artedi 5C T16452 I221 F9 41 Coregonus clupeaformis 5D T16451 I221 F8 42 Coregonus artedi 5F T16448 I221 F5 43 Coregonus artedi 5H T16449 I221 F6 44 no barcode 5I T16460 I221 G8 45 Coregonus artedi 5J T16450 I221 F7 46 Etheostoma exile 9A T16453 I221 G1 47 Etheostoma exile 9B NA - - 48 Perca flavescens 9C T16461 I221 G9 49 Etheostoma exile 9D T16455 I221 G3 50 Etheostoma exile 9E T16456 I221 G4 51 Osmerus mordax 10A lost - - 52 Etheostoma exile 10B T16457 I221 G5 53 no barcode 10C T16462 I221 H1 54 Notropis buchanani 10D T16459 I221 G7 55 Etheostoma exile 10E T16458 I221 G6

110 Chapter 4. Extending DNA barcoding coverage for Lake Whitefish (Coregonus clupeaformis) across the three major basins of Lake Huron

4.1. Abstract DNA barcoding is a useful tool for both species identification and discovery, but the latter requires denser sampling than typically used in barcode studies. Lake Whitefish

(Coregonus clupeaformis) is a valuable species, fished traditionally, commercially, and recreationally in Lake Huron. Based on the natural geographic and bathymetric separation of the three major basins in Lake Huron, the potential separation of Lake

Whitefish within these basins, and the variation among life history (early and late spawning), we predicted that Lake Huron might harbour cryptic lineages of Lake

Whitefish at the basin level. To test this prediction, DNA barcodes of the mitochondrial

5’ cytochrome c oxidase subunit I (COI) gene sequences were recovered from spawning- phase Lake Whitefish (n=5 per site), which were collected from sites (n=28) around Lake

Huron during Fall 2012. These sequences, combined with other publically available

DNA barcodes from the Barcode of Life Data System (BOLD), revealed twelve unique haplotypes across North America, nine haplotypes in Lake Huron with seven unique. The dominant haplotype was found throughout Lake Huron and east to the St. Lawrence

River. No deep divergences were revealed. This comprehensive lake-wide sampling effort offers a new perspective on C. clupeaformis and can provide insight for environmental assessments and fisheries management.

______

Accepted to the Topical Issue of DNA Barcodes – Barcoding of Fishes and co-authored by H. E. Braid, S. S. Crawford and R. H. Hanner

111 4.2. Introduction

Fish identification can be challenging because of ontogenetic variation and a limited understanding of taxon boundaries (Serrao et al. 2014). DNA barcoding is a tool for species identification (Hebert et al. 2003) that applies to all life stages. It relies on the existence of species-specific differences found in short standardized mitochondrial 5’ cytochrome c oxidase subunit I (COI) gene sequences and has successfully been used for species discrimination in fishes (e.g., Ward et al. 2009; Hubert et al. 2008; Hubert et al.

2010). As a tool, barcoding lies at the interface of population genetics and phylogenetics

(Hajibabaei et al. 2007). Although DNA barcodes are primarily used for species identification, they also supports species discovery in the well-studied North American fauna by revealing deeply divergent genetic lineages that may represent cryptic species

(April et al. 2011; Young et al. 2013). In order to appropriately apply DNA barcoding as a tool for species identification and discovery, there is a need to improve our knowledge of genetic variation both within and between species. This variation is important because species-level identification generally relies on the existence of a “barcode gap”, which is a separation between the maximum intraspecific variation versus the minimum interspecific distance (Meier et al. 2008; Knebelsberger et al. 2014).

The Fish Barcode of Life (FISH-BOL) initiative has focused on generating DNA barcodes for all fish species globally (Ward et al. 2009; Becker et al. 2011). DNA barcoding has proven useful in identifying North American freshwater fishes (Hubert et al. 2008; Becker et al. 2011; April et al. 2013), with a 95% coverage of barcodes for

Canadian freshwater fish species (Hubert et al. 2008). Although a sample size of five individuals per species was recommended at the commencement of FISH-BOL (Hanner

112 et al. 2005), there have been more recent suggestions that greater spatial coverage and increased sample size per species are necessary to accurately represent intraspecific genetic variation (Young et al. 2013).

Lake Whitefish, Coregonus clupeaformis, is a culturally and socio-economically important species in the Laurentian Great Lakes. Lake Huron is comprised of three major basins (Main Basin, North Channel and Georgian Bay) and is the second largest of the

Great Lakes and the fourth largest lake in the world (Beeton & Saylor 1995). Within the lake, the three major basins are partially separated bathymetrically by ridges and channels

(Sloss & Saylor 1975; Bohm 1985). Lake Whitefish are found throughout the lake, with a heterogeneous distribution of spawning grounds through each of the basins. There have been some reports at some spawning grounds of different phases of whitefish spawners through the spawning season (e.g. early and late spawning stages), typically extending from October to December, with phenotypes that have been distinguished by First

Nations and harvesters (Loftus 1980; Stott et al. 2012). For example, interviews with

Lake Whitefish commercial fishermen suggested that when fishing for Lake Whitefish in

Lake Huron there was a chance of seeing up to three different morphologies of Lake

Whitefish; some forms were long and not so deep, others were more rounded (Loftus

1980). It has also been suggested that smaller-sized whitefish are present in the south end of Georgian Bay, while the more northern fish were ‘thicker’ (Loftus 1980). DNA barcodes have shown in many cases that subtle morphological variation is correlated with barcode divergence; corroboration from independent data partitions highlights the existence of cryptic diversity (Mat Jaafar et al. 2012; Hubert et al. 2012; Winterbottom et al. 2014). The variation in morphology and spawning chronology, combined with the

113 inherent challenges associated with whitefish taxonomy generally, suggest the possibility of cryptic Lake Whitefish lineages between and/or within basins.

In previous studies, DNA barcoding has been effective in discriminating Lake

Whitefish from other species (i.e. April et al. 2013). However, the results should be considered preliminary due to limited sample sizes. Increased sampling would be needed to detect potential cryptic lineages as previously demonstrated in an investigation of sculpins (Cottus spp.) in a North American riverine system (Young et al. 2013). To date, there were few Lake Whitefish barcodes available on the Barcode of Life Data system

(BOLD; Ratnasingham & Hebert 2007) with little representation of possible haplotype variation. The goal of this study was to investigate basin-level haplotype variation and potential cryptic diversity of Lake Whitefish in Lake Huron, by expanding the barcode library across known spawning sites and times. This study is novel in that it represents the first comprehensive barcode sampling effort for a species inhabiting a large and bathymetrically complex lake ecosystem.

4.3. Materials and Methods

Specimen Collection

A lake-wide sampling effort for adult Lake Whitefish was undertaken in the fall of 2012 in Lake Huron (Figure 4.1; Table 4.1). Spawning-phase Lake Whitefish were sampled from twenty-eight locations across the three basins of Lake Huron: the Main

Basin (n=16), the North Channel (n=4), and Georgian Bay (n=8). Early and late samples were taken from Sites 16-18 (Cape Hurd, Stokes Bay, and Howdenvale) along the

Western shore of the Saugeen (Bruce) Peninsula based on known variation in spawning

114 activity of Lake Whitefish in these areas (Table 4.1). Five different individuals, including at least two males and two females, were sampled from each site for DNA barcoding.

Two muscle fillets were taken from each individual and maintained at -20oC, thawed briefly for subsampling of tissue of approximately 1 gram of tissue for DNA extraction.

We had previously obtained representative Lake Whitefish from Stokes Bay and Sharbot

Lake, which were processed in the same manner.

DNA Barcoding

DNA was extracted from the selected individuals using Xytogen Animal

Extraction Kits (Xytogen, Perth, Australia) following manufacturing guidelines. DNA was diluted with ddH2O to 1:10 before use in PCR (~50 ng/µl). The 652 bp “barcode” region of COI (Hebert et al. 2003) was amplified using universal fish primers VF1i_t1

(TGT AAA ACG ACG GCC AGT TCT CAA CCA ACC AIA AIG AIA TIG G) and

VR1i_t1 (CAG GAA ACA GCT ATG ACT AGA CTT CTG GGT GIC CIA AIA AIC

A) (Ivanova et al. 2007). PCR amplification was carried out in 12.5 µL reaction volumes with 6.25 µL 10% trehalose, 2 µL ddH2O, 1.25 µL 10X buffer, 0.625 µl MgCl2 (50 mM),

0.1 µL VF1i_t1 (10 µm), 0.1 µL VR1i_t1 (10 µm), 0.0625 µL 10 mM dNTPs, 0.06 µL

Platinum Taq polymerase (5 U/µL), and 2 µL of DNA. The reaction profile was: an initial hot start at 94°C for 120sec; followed by forty cycles with a denaturation at 94°C for 30 sec, annealing temperature of 52°C for 40 sec, and extension at 72°C for 60sec; followed by a final extension at 72°C for 10min and hold at 4°C indefinitely.

Amplification success of PCR products was ascertained visually using 2% Agarose E- gels (Invitrogen), where a clear band on the gel indicated successful amplification.

115 Sequencing reactions for PCR products used BigDye v3.1 and the same primers used for amplification, and sequenced bidirectionally. Sequences were edited using

Sequencher version 4.9 (Gene Codes Corporation, Ann Arbor, MI, USA) and aligned using MEGA 5.2 (Tamura et al. 2007). Sequences were uploaded to the Barcode of Life

Database (BOLD; Ratnasingham & Hebert 2007) under the container project “Lake

Whitefish (Coregonus clupeaformis)” (project code: LOLW), in the public project “Lake

Wide Lake Huron Lake Whitefish” (project code: LWLHW; accession numbers

KP978068- KP978225). These were combined with other sequences for Lake Whitefish that we generated from Sharbot Lake (60 sequences, project code: SLHL; accession numbers KP978226 - KP978312) and Stokes Bay (28 sequences, project Stokes Bay,

Ontario, Lake Whitefish (project code: SBOLW; accession numbers KP978018-

KP978067). Other sequences were obtained from public projects available on BOLD: eight sequences from locations across Canada were taken from the project Barcoding of

Canadian Freshwater Fishes (project code: BCF). Together, these sequences were combined into a single “data set” (DS-CORG) on BOLD and given a unique digital object identification number (http://dx.doi.org/10.5883/DS-CORG). Novel sequences from this study were also submitted to GenBank (accession numbers KP978068-

KP978225).

Haplotype Analysis

A haplotype network was generated for the haplotypes of Lake Whitefish using a median-joining network (Bandelt et al. 1999) constructed in Network 4.6.1.2 using default settings (www.fluxus-engineering.com). Distribution maps of haplotypes were

116 created using ArcGIS 10.2 (Environmental Systems Research Institute [ESRI], Redlands,

CA). The map of DNA barcode haplotypes from across North America, excluding the dominant haplotype A, includes a shapefile of ecoregions based on Abell et al. (2008) from http://www.feow.org. The Lake Huron coastline data were mapped using GIS layers (glgis_gl_shore_noaa_70k and lake_huron_bathymetry) published by the Great

Lakes Information Network (http://gis.glin.net/). The size of each pie is proportional to the number of individuals, and each colour symbolizes a different haplotype (Figure 4.1).

Sample sizes are proportional in Figure 4.1, except for Lake Huron due to the large number of haplotype A, which would have obscured the figure.

4.4. Results

A total of 208 Lake Whitefish sequences were analyzed, associated with a single

Barcode Index Number (BIN) for all known specimens, yielding a total of 12 different haplotypes. The BIN system forms clusters of barcode sequences that can document taxonomic diversity (Ratnasingham & Hebert 2013). From the lake-wide Lake Huron sampling, full-length DNA barcodes were successfully obtained for 148 out of the 154

(96.1%) specimens representing nine haplotypes, seven of which were unique to Lake

Huron (Figure 4.1, 4.2). In eastern North America, one dominant haplotype (A, dark blue) representing 91.8% (191 individuals) of the sequences analyzed appears to be the most predominant, with occurrences throughout Lake Huron and east to the St. Nicolas

River near the St. Lawrence River (Figure 4.2).

Four additional haplotypes were found in locations across Canada. Two individuals from the St. Nicholas River represented the unique haplotype J (light blue),

117 which had a single base pair different from the dominant haplotype A. Three haplotypes were found that differed from the dominant haplotype A by two base pairs: three individuals from Swan Lake, British Columbia represented haplotype G (teal); two unique haplotypes were identified from single individuals from the Yukon River in the

Yukon Territory for haplotype H (grey) and I (brown).

In Lake Huron, the seven unique haplotypes occurred at low frequencies, with each differing from the dominant haplotype A by only a single base pair. Haplotype, B

(dark green), was represented by three individuals from outer Saginaw Bay (Main Basin),

North Island (Main Basin), and Bedford Island (North Channel). Haplotype C (light green) represented two individuals, one respectively from Stokes Bay and Loscombe

Bank (Main Basin). There were five additional singleton unique haplotypes found in

Lake Huron: Haplotype E (red), on the eastern shores of the Main Basin; Haplotypes K

(pink) and L (purple) found in Stokes Bay on the eastern side of Main Basin; Haplotype F

(yellow) on the western shores of the Main Basin; and Haplotype D (orange) in the North

Channel. Of the three sites that were sampled in early and late spawning periods, both

Stokes Bay (Site 17) and Howdenvale (Site 18), respectively, exhibited one unique haplotype in the late spawning phase along with the dominant haplotype, while Cape

Hurd (Site 16) did not show any unique haplotypes. The dominant haplotype was found across all spawning sites at all sampling periods (Figure 4.2B).

4.5. Discussion

After extensive lake-wide sampling of individuals of spawning-phase Lake

Whitefish from Lake Huron, very little variation was found in the DNA barcodes,

118 providing no evidence for basin-level cryptic lineages, consistent with a single species

(e.g. no disconnected haplotype networks, Hart & Sunday 2007). Haplotype A, which was dominant in the entire collection, was also dominant throughout all three basins of

Lake Huron, including both the early and late spawning stages. In contrast, haplotype B

(dark green) was found only in Lake Huron in a limited number of locations. Haplotype

B was found in both Saginaw Bay (Main Basin) on the south western coast of Lake

Huron and north of Douglas Point (Main Basin) on the east side of the bay. Additionally, haplotype B was found in the north side of Manitoulin Island (North Channel). These patterns of haplotype distributions do not suggest cryptic lineages (Spangler 1970;

Carmichael & Deary 2009).

Our comprehensive lake-wide sampling of Lake Whitefish barcodes does not provide evidence for the existence of cryptic species among three main basins of Lake

Huron. Of the three sites that were sampled in both early and late spawning stages, two sites were found to have rare haplotypes in the late stage, but those individuals were always found to co-occur with individuals that had the dominant haplotype. The occurrence of unique haplotypes among late stage spawners may suggest a temporal, rather than a spatial trend in cryptic lineages; however, due to the co-occurrence of the dominant haplotype, this seems unlikely.

The goal of this study was to investigate basin-level haplotype variation and examine the potential for cryptic diversity within Lake Whitefish in Lake Huron. While new haplotype variants were discovered at some sites (rare and in low frequencies), these haplotypes were only 1-2 nucleotides different from the dominant haplotype and always co-occurred with the dominant haplotype at the same collection site/time. From the

119 substantial expansion of the Lake Whitefish barcode library, no cryptic diversity (e.g. deeply divergent lineage) was noted in Lake Huron and no basin-specific haplotype variation was revealed. This study represents one of the most comprehensive barcode surveys of an individual fish species ever undertaken and stands in contrast to Young et al. (2013) riverine system, which discovered substantial cryptic diversity with increased sampling effort.

This study fails to support the hypothesis that cryptic lineages of Lake Whitefish exist in Lake Huron, and, although it does reveal the presence of rare barcode haplotypes that seem to be unique to specific sampling sites, it does not address population structure.

Such investigations require more sensitive methods and, ideally, sampling more individuals from each putative population, including individuals that were morphologically identified as different. It is important to note that population discrimination of Lake Whitefish in Lake Huron has been identified as a key ecological uncertainty for fisheries management and environmental assessments (Casselman et al.

1981; Crawford et al. 2001; Stott et al. 2012) and should be the focus of future work.

However, further sampling of Lake Whitefish among the Laurentian Great Lakes and throughout the species range may yet provide evidence for cryptic whitefish diversity elsewhere and should be a topic for further exploration. Extensive sampling is a necessary step in the development of a robust species-specific PCR identification assay for any given species of interest (Bustin 2005). Hence, this study paves the way for developing a more cost-effective method of identifying whitefish specimens, which might be a concern because of the species’ status as a valued ecosystem component.

120 4.6. References Abell R., Thieme M.L., Revenga C., Bryer M., Kottelat M., Bogutskaya N., et al. 2008. Freshwater ecoregions of the world: a new map of biogeographic units for freshwater biodiversity conservation, BioScience 58(5): 403-414.

April J., Mayden R.L., Hanner R.H., Bernatchez L. 2011. Genetic calibration of species diversity among North America’s freshwater fishes. PNAs 108(26): 10602-10607

April J., Hanner R.H., Dion-Cote A-M., Bernatchez L. 2013. Glacial cycles as an allopatric speciation pump in north-eastern American freshwater fishes. Molecular Ecology 22(2): 409-422.

Bandelt, H-J., Forster P., Rohl A. 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16(1): 37-48.

Beeton A.M., Saylor J.H., Limnology of Lake Huron. In: Munawar M., Edsall T., Leach J. 1995 (eds). The Lake Huron Ecosystem: Ecology, Fisheries and Management. SPB Academic Publishing, Amsterdam, The Netherlands. p 1-37.

Becker S., Hanner R., Steinke D. 2011. Five years of FISH-BOL: Brief Status Report. Mitochondrial DNA 22(S1): 3-9.

Bohm E.U. 1985. Lake Huron Nearshore Currents in the Vicinity of Bruce NPD 1983- 1984, Geotechnical and Hydraulic Engineering Department Report No 85301, 1-54.

Bustin, S.A. 2005. Real-Time PCR. Encyclopedia of Diagnostic Genomics and Proteomics. University of London, London, U.K., 1117-1125.

Carmichael K., Deary C. 2009. Filling a data gap- lake whitefish (Coregonus clupeaformis) index netting in the North Channel of Lake Huron. Aquatic Ecosystem Health & Management 12: 23-28.

Casselman J.M., Collins J.J., Crossman E.J., Ihssen P.E., Spangler G.R. 1981. Lake whitefish (Coregonus clupeaformis) stocks of the Ontario waters of Lake Huron. Canadian Journal of Fisheries and Aquatic Science 38: 1772-1789.

Crawford S.S., Muir A., McCann K. 2001. Ecological basis for recommendation of 2001 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron, report prepared for the Chippewas of Nawash First Nation, Wiarton, Ontario. 156pp.

Hajibabei M., Singer G.A.C., Hebert P.D.N., Hickey D.A. 2007. DNA barcoding: how it complements taxonomy, molecular phylogenetics and population genetics. TRENDS in Genetics 23(4): 167-172.

121 Hanner R., Schindel D., Ward B., Hebert P. 2005. Fish barcode of life (FISH-BOL). 1– 19.

Hart M.W., Sunday J. 2007. Things fall apart: biological species form unconnected parsimony networks. Biological Letters 3(5): 509-512.

Hebert P.D.N., Cywinska A., Ball S.L., DeWaard J.R. 2003. Biological identifications through DNA barcodes. Proc of the Royal Society of London. 270: 313-321.

Hubert N., Hanner R., Holm E., Mandrak N.E., Taylor E., Burridge M., et al. 2008. Identifying Canadian Freshwater Fishes through DNA Barcodes. PLoS One 3(6): e2490.

Hubert N., Delrieu-Trottin E., Irisson J-O., Meyer C., Planes S. 2010. Identifiying coral reef fish larvae through DNA barcoding: a test case with the families Acanthuridae and Holocentridae. Molecular Phylogenetics and Evolution 55: 1195-1203.

Hubert N., Meyer C.P., Bruggemann H.J., Guerin F., Komeno R.J.L., Espiau B., et al., 2012. Cryptic diversity in Indo-Pacific Coral-Reef fishes revealed by DNA-Barcoding provides new support to the centre-of-overlap hypothesis. PLoS One 7 (3): e28987.

Ivanova N.V., Zemlak T.S., Hanner R.H., Hebert P.D.N. 2007. Universal primer cocktails for fish DNA barcoding. Molecular Ecology Notes 7: 544-548.

Knebelsberger T., Dunz A.R., Neumann D., Geiger M.F., Molecular diversity of Germany’s freshwater fishes and lampreys assessed by DNA barcoding. Molecular Ecology Resources 2014, 1-11.

Loftus D.H. 1980. Interviews with Lake Huron commercial fishermen, Ontario Ministry of Natural Resources, Lake Huron Fisheries Assessment Unit Report 1-80. Owen Sound, Ontario, 633pp.

Mat Jaafar T.N.A., Taylor M.I., Mohd Nor S.A., de Bruyn M., Carvalho G.R. 2012. DNA barcoding reveals cryptic diversity within commercially exploited Indo-Malay Carangidae (Teleosteii: Perciformes). PLoS One 7(11): e49623.

Meier R., Zhang G., Ali F. 2008. The use of mean instead of smallest interspecific distance exaggerates the size of the ‘barcoding gap’ and leads to misidentification. Systematic Biology 57: 809–813.

Ratnasingham S., Hebert P.D.N. 2007. BOLD: The Barcode of Life Data System (http://www.barcodinglife.org). Molecular Ecology Notes 7(3): 355-364.

Ratnasingham S., Hebert P.D.N. 2013. A DNA-Based registry for all animal species: The Barcode Index Number (BIN) System. PLoS One 8(7): e66213.

122 Serrao N.R., Steinke D., Hanner R.H. 2014. Calibrating snakehead diversity with DNA barcodes: Expanding taxonomic coverage to enable identification of potential and established invasive species. PLoS One 9(6): e99546.

Sloss P.W., Saylor J.H. 1975. Measurements of current flow during summer in Lake Huron. U.S. Department of Commerce, National Oceanic and Atmospheric Administration Environmental Research Laboratories ERL 353-GLERL 5, 1-45.

Spangler G.R. 1970. Factors of mortality in an exploited population of whitefish, Coregonus clupeaformis, in northern Lake Huron. In: Lindsey C.C., Woods C.S. (eds), Biology of Coregonid Fishes, University of Manitoba Press, Winnipeg, Manitoba, 515- 529.

Tamura K., Dudley J., Nei M., Kumar S. 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24: 1596-1599.

Stott W., Ebener M.P., Mohr L., Schaeffer J., Roseman E.F., Harford W.J., et al. 2012. Genetic structure of lake whitefish, Coregonus clupeaformis, populations in the northern main basin of Lake Huron. In: Tallman R.F., Howland K.L., Rennie M.D., Mills K. (eds) Biology and management of coregonid fishes – 2008; 10th International Symposium on the Biology and Management of Coregonid Fishes, Winnipeg, Canada. Advances in Limnology 63: 241-260.

Ward R.D., Hanner R., Hebert P.D.N. 2009. The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology 74: 329-356.

Winterbottom R., Hanner R.H., Burridge M., Zur M. 2014. A cornucopia of cryptic species – a DNA barcode analysis of the gobiid fish genus Trimma (Percomorpha, Gobiiformes). Zookeys 381: 79-111.

Young M.K., McKelvey K. S., Pilgrim K. L., Schwartz M. K. 2013. DNA barcoding at riverscape scales: assessing biodiversity among fishes of the genus Cottus (Teleostei) in northern Rocky Mountain streams. Molecular Ecology Resources 1-13.

123

Figure 4.1. DNA barcode haplotype variation found in Lake Whitefish (Coregonus clupeaformis) across North American ecoregions. Each colour represents a different haplotype. Haplotypes were included from Yukon River, Swan Lake, St. Laurence River, Sharbot Lake, and Lake Huron. The size of each pie chart is proportional to the sample size from each location; however, the most dominant haplotype (A) has been excluded from Lake Huron for the purpose of visualization of the less prevalent haplotypes, but all haplotypes for Lake Huron can be seen in Figure 4.2.

124

Figure 4.2. DNA barcode haplotype variation found in Lake Whitefish (Coregonus clupeaformis); site numbers correspond to Table 4.1 and L1 and L2 are two sites sampled in BOLD project Stokes Bay, Ontario, Lake Whitefish [SBOLW]. (A) Haplotype network analysis of DNA barcode sequences from across North America; the size of the nodes corresponds to the number of individuals that share each haplotype; the colour of each unique haplotype corresponds to the pie charts in B; (B) Geographic distribution of haplotypes from Lake Huron at each site sampled during 2012, the size of each pie chart is proportional to the sample size from each location. MB=Main Basin, GB=Georgian Bay, NC= North Channel.

125 Table 4.1. Collection site data for the Lake Whitefish (Coregonus clupeaformis) for the 2012 sampling effort in Lake Huron. Map number corresponds to the site number indicated on Figure 1; Basin name abbreviations are NC = North Channel, GB=Georgian Bay, and MB= Main Basin; the sex of fish sampled is indicated by M (male) and F (female); early designated the start of the fall spawning season, late designates the end of the fall spawning season.

Site name Map Basin Latitude Longitude Number of number fish sampled Thessalon 1 NC 46.23 -83.43 2M, 3F South Cockburn 2 MB 45.87 -83.46 3M, 2F Blind River 3 NC 46.16 -82.92 3M, 2F Vidal Bay 4 NC 45.94 -83.09 3M, 2F Burnt Island 5 MB 45.8 -82.46 2M, 3F Providence Bay 6 MB 45.64 -82.31 2M, 3F Bedford Island 7 NC 46.01 -82.07 2M, 3F Wiki-Smith Bay 8 GB 45.86 -81.63 2M, 3F Bad River 9 GB 45.91 -80.99 2M, 3F Henvey Inlet 10 GB 45.84 -80.73 3M, 2F Christian Island 11 GB 44.88 -80.16 2M, 2F Maurice Point 12 GB 44.72 -80.07 2M, 3F Meaford 13 GB 44.62 -80.53 2M, 3F Cape Croker South 14 GB 44.85 -80.94 3M, 2F Sydney Bay 15 GB 44.91 -81.07 2M, 2F Cape Hurd 16 MB 45.18 -81.75 2M, 3F early 3M, 2F late Stokes Bay 17 MB 45.02 -81.48 3M, 2F early 3M, 2F late Howdenvale 18 MB 44.85 -81.35 1M, 4F early 3M, 2F late Scougall Bank 19 MB 44.39 -81.53 2M, 3F Kettle Point 20 MB 43.20 -82.04 3M, 2F Sarnia 21 MB 43.03 -82.31 2M, 3F Outer Saginaw Bay 22 MB 44.15 -83.08 2M,2F North Island 23 MB 43.87 -83.44 3M, 1F East Tawas 24 MB 44.27 -83.46 3M, 2F Alpena 25 MB 44.86 -83.25 3M, 2F North Point 26 MB 45.03 -83.24 3M, 2F Hammond Bay 27 MB 45.50 -84.03 2M, 3F Search Bay 28 MB 45.98 -84.50 2M, 3F

126 Chapter 5. Real-Time PCR identification of Lake Whitefish (Coregonus clupeaformis) in the Laurentian Great Lakes.

5.1. Abstract

Ichthyoplankton, especially those in the subfamily Corigoninae, can be very difficult to identify due to their small size and lack of distinguishing meristic and morphometric characteristics. Coregonines in the Laurentian Great Lakes are socially and economically important to both Indigenous and non-Indigenous people, and there is an urgent need to develop a technique for rapid and accurate species identification for environmental and fisheries assessments. Both visual identification and DNA barcoding can be time consuming and costly when dealing with large volumes of ichthyoplankton samples. Real-time PCR is emerging as a useful tool for rapid species identification because it requires no post PCR processing (e.g. gels). The purpose of this study was to develop a real-time PCR assay to specifically identify Lake Whitefish (Coregonus clupeaformis) in larval fish assemblages based on a 122bp amplicon. The efficiency of the reaction, as calculated from the standard curve, was 90.77% with the standard curve having an r-squared value of 0.998. Specificity of the assay provided only one melt peak in a melt- curve analysis and amplification of only the target species. The assay successfully identified target DNA in as low as 0.01% proportion of a DNA mixture. This assay was designed on the portable Smart Cycler II platform and can be used in both field and laboratory settings to successfully identify Lake Whitefish.

______

In review in Journal of Fish Biology co-authored by H. E. Braid, A.M. Naaum, S. S. Crawford and R. H. Hanner

127 5.2. Introduction

Species-level identification of ichthyoplankton can be extremely challenging due to the small size of larvae and lack of distinguishing morphometric and meristic characteristics (Scott & Crossman 1973; Teletchea 2009; Chapter 3). The proper identification of ichthyoplankton is extremely important in fish ecology, especially for understanding important spawning and nursery habitats, life history dynamics, and year- class forecasting in fishery management (Ko et al. 2013). The classical method for identification of larval fish relies on morphology-based keys (e.g., Scott & Crossman

1973; Ko et al. 2013), which can be highly unreliable and inaccurate (Hare et al. 1994;

Kochzius et al. 2008).

The morphological identification of larval fish in the subfamily Coregoninae is particularly challenging because phenotypic characteristics often vary across environments, and there are few distinct morphological characteristics for species with overlapping ranges (Scott & Crossman 1973; Auer 1982; Schlei et al. 2008). In the

Laurentian Great Lakes, coregonines such as Lake Whitefish Coregonus clupeaformis,

Lake Herring C. artedi, Bloater C. hoyi, Shortjaw Cisco C. zenithicus and Round

Whitefish Prosopium cylindraceum, are important species socially, commercially and ecologically to both Indigenous and non-indigenous human communities (Crawford et al.

2001; Golder Associates 2008). Within the Great Lakes, other coregonids such as the deepwater Ciscos C. kiyi and C. johannae, the Shortnose Cisco C. reighardi, the Blackfin

Cisco C. nigrippinus and the Longjaw Cisco C. alpenae have been reported, with C. alpenae, C. nigrippinis, C. reighardi and C. johannae reported as being extinct (Todd &

Smith 1992; Davis & Todd 1998; Roth et al. 2012). Coregonus kiyi and C. nigripinnis

128 are reported as extirpated, and C. lavaretus and C. moraena were introduced into Lake

Huron but failed to establish (Roth et al. 2012). Among the commercially harvested species, larvae of C. artedi and C. clupeaformis have previously been reported as very difficult to differentiate on the basis of morphological characteristics (Todd & Stedman

1980). Both C. clupeaformis and P. cylindraceum have been specifically classified as

‘Valued Ecosystem Components’ (VEC) in Lake Huron because of their ecological significance, economic value, and presence of important spawning locations in Lake

Huron (Holmes & Noakes 2002; Brown 2007; Brown 2008). For the purpose of this proof-of-concept study, we focus on Lake Whitefish, which is the most intensively harvested species in the Lake Huron Indigenous and non-Indigenous fisheries (Mohr &

Nalepa 2005; LaRiviere & Crawford 2013).

Visual identification of larval Lake Whitefish relies on morphological keys, notably those prepared by Auer (1982) and Cucin and Faber (1985). However, such visual identification can be costly, labour intensive and unreliable for samples that contain multiple individuals and/or species (Pfrender et al. 2010). In addition, morphological species identification of larval samples can be particularly difficult because individuals may be badly damaged by sampling methods, and often only fragmentary remains are recovered (Ager et al. 2006). Genetic techniques, including

DNA barcoding and restricted fragment length polymorphisms (RFLPs), have the potential to greatly increase accuracy and reliability of larval fish identification (e.g.,

Kochzius et al. 2010; Ko et al. 2013) as well as the identification of fragmentary remains

(Quinteiro et al. 1998; Mackie et al. 1999). Several genetic methods have previously been applied to identify C. clupeaformis, including: allozymes to identify Lake Whitefish

129 populations (Casselman et al. 1981), microsatellites to infer glacial lineages (Lu et al.

2001), RFLPs to evaluate genetic diversity and geographic structure (Bernatchez &

Dodson 1991), and DNA barcoding for species identification (Hubert et al. 2008).

Although these genetic techniques are effective in identifying C. clupeaformis, and provide some benefits over visual identification, they can still be very costly and time consuming for large-scale implementation. For these reasons, it is necessary to develop novel genetic methods for reliably identifying C. clupeaformis in large, mixed samples of wild ichthyoplankton.

Real-time PCR (qPCR) has been proposed as an alternative method for species identification to address the issues of cost, time, and handling of multispecies batch samples like the mashes resulting from ichthyoplankton surveys (Bustin 2005). Real- time PCR has proven to be a reliable method for species identification in many fields, including pest identification (Huang et al. 2010; Naaum et al. 2012) and seafood market fraud (Taylor et al. 2002; Rasmussen Hellberg et al. 2011). This method has also been applied successfully in several aspects of fish ecology including the identification of fish eggs (Taylor et al. 2002; Bayha et al. 2008; Gleason & Burton 2011); the identification of marine fish parasites (McBeath et al. 2006); and the differentiation of commercially important salmon and trout species (Rasmussen Hellberg et al. 2010). Real-time PCR can also allow for relative quantification of individual larvae in ichthyoplankton samples

(Pan et al. 2008). In terms of technical advantages, real-time PCR identification can be performed rapidly, producing results in as little as one hour, and requires no post-PCR processing step(s) including gel electrophoresis; real-time PCR can also be performed onsite with portable platforms, providing enhanced flexibility and responsiveness for

130 field assessments of species identification (Naaum et al. 2012). This makes real-time

PCR an ideal alternative to current methods for species identification of Lake Whitefish.

In this study, genus-specific PCR primers and species-specific probe for real-time

PCR were designed for C. clupeaformis using the DNA barcode region. DNA barcoding, a method of species identification that utilizes sequence variation in the mitochondrial cytochrome c oxidase subunit 1 (COI) gene to discriminate species (Hebert et al. 2003), has proven to be particularly useful for fishes (Ward et al. 2009) including those from

North America (Hubert et al. 2008; April et al. 2012). Larval fish have been successfully identified using DNA barcoding in Australia (Pegg et al. 2006) among coral reefs in the

Indo-Pacific (Hubert et al. 2010); along the Caribbean coast of Panama (Victor et al.

2009); and along the Yucatan Peninsula, Mexico (Valdez-Moreno et al. 2010). The DNA barcode region was used in the development of this assay because of this regions’ similarity within species along with difference between-species, because of good quality sequences publically available on the Barcode of Life Data Systems (BOLD)

(Ratnasingham & Hebert 2007) and because barcoding (e.g. Sanger sequencing) could be used to verify species identifications.

5.3. Material and Methods

Sample Collection and DNA Extraction

Larval samples were provided by third parties as preserved tissues in 95% ethanol. Larval specimens were first examined by visual identifiers. Caudal fin clips were taken from each larval specimen. Two samples, Atlantic Salmon (Salmo salar) and

Round Whitefish (Prosopium cylindraceum) were collected from adult muscle samples.

131 DNA was extracted using the Qiagen DNeasy Blood and Tissue Kits (Qiagen, Valencia,

CA) following the manufacturer’s instructions to confirm identification of samples to validate this study.

DNA Barcoding

PCR amplification for the DNA barcode region was carried out in 12.5 µl reaction volumes with 6.25 µl 10% trehalose, 2 µl ddH2O, 1.25 µl 10X buffer, 0.625 µl MgCl2

(50 mM), 0.1 µl each of forward and reverse 10 µM primers (universal fish primers

VF1i_t1 and VR1i_t1; Ivanova et al. 2007). 0.0625 µl 10 mM dNTPs, 0.06 µl Platinum

Taq polymerase (5 U/µl), and 2 µl of DNA. The reaction profile was: an initial hot start at 94°C for 120 sec; followed by forty cycles with a denaturation at 94°C for 30 sec, annealing temperature of 52°C for 40 sec, and extension at 72°C for 60 sec; followed by a final extension at 72°C for 10min and hold at 4°C indefinitely. Amplification success of PCR products was ascertained visually using pre-cast 2% Agarose E-gels (Invitrogen), where a single, distinct, unambiguous band on the gel indicated successful amplification.

DNA sequencing reactions were carried out in 14 µl reaction volumes with 1 µl

BigDye v3.1, 1 µl 5X SeqBuffer, 1 µl primer [10 µM] (C_VF1LFt1 for forward and

C_VR1LRt1 for reverse), 10 µl ddH2O, and 1.5 µl PCR product. PCR products were bidirectionally sequenced using an ABI 3730 DNA Analyzer (Applied Biosystems). Bi- directional sequence contig assemblies were created and edited using Sequencher v. 4.9

(Gene Codes Corporation, Ann Arbor, MI, USA) and multiple sequence alignments were generated manually using MEGA 5.2 (Tamura et al. 2007). Sequences were uploaded to the Barcode of Life Data System (BOLD; Ratnasingham & Hebert 2007) public project

132 “Stokes Bay, Ontario, Lake Whitefish” (project code: SBOLW) and subsequently submitted to GenBank (accession numbers KP978018-KP978067) via BOLD. All barcodes were queried against the sequences of known provenance using the BOLD ID engine (www.boldsystems.org, Ward et al. 2009) using the “Species ID” option. A DNA barcode species match was defined as a percent of sequence similarity greater than 99% to a unique species in the reference database.

Real-time PCR

The primers and probe were designed using all North American Coregoninae

DNA barcodes available publically in BOLD, including sequences generated for this study (including Sharbot Lake Hatchery Larvae; Stokes Bay, Ontario, Lake Whitefish; and Lake Wide Lake Huron Lake Whitefish) and sequences available on GenBank (Table

5.1). European species were excluded from probe development (Table 5.2). A dataset entitled “Dataset for Coregonids” (DS-CORG), given a unique digital object identification number (http://dx.doi.org/10.5883/DS-CORG), was created in BOLD with all related species sequences and other Great Lakes fish sequences from the container project “Lake Whitefish (Coregonus clupeaformis)” (project code: LOLW), the public projects “Sharbot Lake Hatchery Larvae” (project code: SLHL; accession numbers

KP978226 - KP978312), “Stokes Bay, Ontario, Lake Whitefish” (project code: SBOLW; accession numbers KP978018-KP978067), and “Lake Wide Lake Huron Lake Whitefish”

(project code: LWLHW; accession numbers KP978068- KP978225). Available DNA barcodes were collapsed into unique haplotypes using DNA Barcoding Tools

133 (http:// www.ibarcode.org; Singer & Hajibabaei 2009). Primers and probe were designed using Allele ID 7.75 (Premier Biosoft International, Palo Alto, California, USA) using default settings. Due to the close similarities among coregonid species and within their species complexes (i.e. Coregonus artedi complex and Coregonus clupeaformis complex), primers were designed to be genus specific instead of species specific (Table

5.3). The C. clupeaformis specific probe was designed and tagged with the fluorescent reporter 6 - carboxyfluorescein (FAM) at the 5’ end and BQH-1 Quencher at the 3’ end

(Integrated DNA Technologies, Coralville, IA, USA). The primer and probe were screened in silico using the primer BLAST search by blasting the probe in the position of one of the primers (either forward or reverse) to determine if it matched any of the species (http://www.ncbi.nlm.nih.gov/tools/primer-blast/; Ye et al. 2012.), with matches found to C. clupeaformis and Coregonus pidschian for the primers and probe. As previously mentioned, forward and reverse primers matched with the genus Coregonus, but did not match with the genus Prosopium or other non-target species tested with this assay (Table 5.3). The species-specific probe did not match non-target species in the genera Prosopium or Coregonus, with the exception of Coregonus pidschian. It is important to note that C. pidschian may cause a false positive with this assay; however, because this species has an Arctic distribution, it should not affect identification of specimens from the Laurentian Great Lakes (Freyhof & Kottelat 2008).

Optimization and initial tests were carried out on a Smart Cycler II (Cepheid,

Sunnyvale, California, USA) platform following manufacturer guidelines. Reactions were carried out in 25 µl volumes containing 17 µl ddH2O, 5 µl template DNA [7.1 ng/

µl], 1.25 µl forward primer [0.5 µM], 1 µl reverse primer [0.4 µM], and 0.75 µl probe

134 [0.3 µM]. OmniMix HS lyophilized mastermix (Cepheid, Sunnyvale, California, USA) was used according to manufacturer guidelines. PCR cycling conditions were an initial hold start at 95oC for 120 seconds, followed by 35 cycles of 95oC for 11 secs, annealing at 62oC for 30 secs, and extension at 72oC for 10 secs for improved specificity. Standard curves were created from 10-fold serial dilutions of C. clupeaformis DNA using DNA from 7.1 ng/ µl to 0.71 pg/ µl to determine efficiency of the assay (Figure 5.1). Ct values obtained from the serial dilutions were then plotted against the logarithm of the template

DNA (ng), and slope (m) was calculated by linear regression. Reaction efficiency (E) for real time PCR was calculated, where E=10(-1/m)-1 (Bustin 2005).

The probe and primer pair for C. clupeaformis was surveyed against larvae of the following non-target species to test specificity: a) five individuals for two congeneric coregines – lake herring (C. artedi) [6.3-14.23 ng/ µl] and bloater (C. hoyi) [7.1 ng/ µl]; b) five individuals for a non-congeneric but spatially overlapping coregonine –Round

Whitefish (Prosopium cylindraceum) [78.1 ng/ µl]; c) single individual of a non- coregonine salmonid – Atlantic Salmon (Salmo salar) [7.1 ng/ µl]; d) single individuals of non-salmonids with spatio-temporally overlapping larvae – Yellow Perch (Perca flavescens) [2.7 ng/ µl], White Sucker (Catostomus commersonii) [5.77 ng/ µl], and

Longnose Sucker (Catostomus catostomus) [12.03 ng/ µl]; and e) single individuals for other non-salmonid Great Lakes fishes found in Lake Huron, which are commonly ecountered– Walleye (Sander vitreus) [7.1 ng/ µl], Rainbow Smelts (Osmerus mordax)

[3.7 ng/ µl], Ghost Shiner (Notropis buchanani) [8.0 ng/ µl], and Iowa Darter

(Etheostoma exile) [4.7 ng/ µl]. The DNA from each of the non-target species was acquired following the aforementioned methods, using DNA extracted from the caudal

135 fin clips of larval specimens previously caught in Lake Huron, with the exception of

Atlantic Salmon DNA, which was acquired from a muscle subsample of a store-bought fillet.

The C. clupeaformis primers and probe pair were also tested on mixtures of DNA from known species obtained from larval specimens of C. clupeaformis, C. hoyi and S. vitreus sampled from the Ontario Ministry of Natural Resources White Lake Fish Culture

Station. The C. clupeaformis primers and probe were tested on the following mixtures:

50% each of C. clupeaformis and C. hoyi DNA; 50% each of C. clupeaformis and S. vitreus; 50% each of C. hoyi and S. vitreus; 33% each C. clupeaformis, C. hoyi and S. vitreus; 17% C. clupeaformis; 10% C. clupeaformis (Table 5.4). Each mixture was made in a volume of 30 µl, with 5 µl aliquoted into each tube for testing. DNA of non-target species was standardized to a concentration of 7.1 ng/ µl, to be consistent with that of C. clupeaformis being used. Ct values were recorded for all samples (Table 5.4).

To assess the utility of this assay on a high-throughput instrument, it was also evaluated using a StepOnePlus Real-Time PCR system (Applied Biosystems). All reactions were carried out using MicroAmp fast 96-well reaction plates (100 µl) (Applied

Biosystems), containing the same volumes and concentrations as the Cepheid

SmartCycler II. To set an appropriate, comparable threshold on the new instrument and based on recommendations from Applied Biosystems User Guide, the fluorescence threshold was manually set to 8000 to ensure it was in the amplification phase based on reference target samples of C. clupeaformis previously run on the Cepheid SmartCycler

II. One replicate of each of the same non-target species and mixtures (as above) were tested, as well as DNA from six individuals of C. clupeaformis.

136 Assay specificity was validated in two ways using melt curve analysis and gel electrophoresis. First, melt curve analysis using PerfeCTa® SYBR® Green FastMix®

(Quanta BioScience Inc., Gaithersburg, MD, USA) was carried out on the Cepheid Smart

Cycler II. Reactions were carried out in 25 µl volumes containing 5.25 µl ddH2O, 5 µl template DNA, 1.25 µl forward primer [0.5 µM], 1 µl reverse primer [0.4 µM], and 12.5

µl of SYBR Green. Four samples were tested including a no-template control (5 µl ddH20), two non-target species (5 µl of C. commersonii and C. catostomus respectively) and one target species (5 µl of C. clupeaformis) to determine the presence of a single melt peak with duplicates of each sample. Specificity was also validated by agarose gel electrophoresis by placing 8 µl of qPCR product in a 2% agarose gel in

Tris/Borate/EDTA (TBE) buffer. Specificity of assay would be indicated by clear, distinct bands for only the target species.

5.4. Results

Using the Cepheid SmartCycler II, the primer and probe set was tested against nine non-target species, including both Coregonus hoyi and C. artedi, which are two closely related species to C. clupeaformis and also found in the Great Lakes. No amplification was observed with C. hoyi, C. artedi, Salmo salar, Catostomus commersonii, C. catostomus or Sander vitreus. Some amplification was observed with non-target species: P. flavescens (Ct=29.66) and E. exile (Ct=31.99). No false negatives were observed, and no signal was observed for no-template “blank” controls when testing the optimal protocol on C. clupeaformis DNA or non-target species. The efficiency of the reaction, as calculated from the standard curve (r2=0.998) was 90.77% (Figure 5.1).

137 The assay successfully identified target DNA in as low as 0.01% proportion of a DNA mixture (Table 5.4).

The protocol, primers and probe set were also evaluated on a StepOnePlus Real- time PCR system, a platform capable of high-throughput analysis. This assay was successfully used to identify the target species, C. clupeaformis, with comparable Ct values to those obtained using the Cepheid SmartCycler II for the same samples (Table

5.5). No false negatives were observed, and no signal was observed for no-template controls. No amplification was observed for non-target species S. salar, C. artedi, O. mordax, C. commersonii, C. catostomus or P. cylindraceum. Some fluorescence output was observed with C. hoyi (Ct =30.31), P. flavescens (Ct=29.89), E. exile (Ct=31.67)

(Table 5.5).

Specificity of melt-curve validation of qPCR primers revealed a single melt peak for C. clupeaformis at 81.8 oC, indicating a single PCR product. This melt peak is comparable to the estimated melting temperature of assay product calculated by Allele ID

(Tm=79.2 oC). No melt peaks were observed in no-template controls or in replicates of non-target species, therefore validating the specificity of the Lake Whitefish forward and reverse primer (Figure 5.2A-D). Specificity was further tested by agarose gel electrophoresis. No-template controls, non-target species and Lake Whitefish were amplified using universal fish cocktail primers (Figure 5.3A) and then compared against qPCR product of no-template control, non-target species and Lake Whitefish using the

Lake Whitefish primers (Figure 5.3B). Bands were present for only Lake Whitefish when visualizing products of qPCR at the expected amplicon length of 122bp for the product of qPCR.

138 5.5. Discussion

A real-time assay was successfully developed that can reliably differentiate between Coregonus clupeaformis and other local coregonines found in Lake Huron

(Table 5.1). Of particular importance is the ability to differentiate larval C. clupeaformis from two other congeneric species, C. hoyi and C. artedi, which can be easily misidentified as C. clupeaformis during the larval period (Scott & Crossman 1973; Auer

1982; Todd & Stedman 1989). The assay correctly identified all target samples of C. clupeaformis, with no false negatives observed. Although some amplification was observed for non-target species, the Ct values were high (Ct=29.66-31.99) compared to the target species (Ct=17.81-22.06) and this was likely due to contamination of DNA samples when larval fish were initially bulk stored in ethanol. With repeat DNA extractions from larval Yellow Perch (Perca flavescens) stored separately in ethanol, nonspecific amplification was eliminated. Three independent samples from larval Yellow

Perch were re-tested with duplication on both the Cepheid SmartCycler II and the

StepOnePlus Real Time PCR system and showed no non-target amplification, with undetected Ct values. This shows that the previous non-target amplification of Yellow

Perch was likely from contaminated DNA and suggests this was also the case for non- target amplification observed in C. hoyi and E.exile, which were collected and stored in the same manner as the previously tested larval Yellow Perch.

The assay developed in this study represents a rapid identification method that can be implemented in the field, or in high-throughput laboratory facilities, as required. The assay is a good complement and/or alternative to visual identification methods, more efficient than other time- and cost-intensive genetic methods of identification, and

139 provides an effective solution to identifying individuals in wild larval fish assemblages.

For example, the time and cost associated with visually identifying a single ichthyoplankton at an estimated cost of $50-100/hr per fish (Greenaway, pers. comm.

Golder Associates. 2012), with DNA barcoding at a cost of $40/sample (R. Hanner, pers. comm. University of Guelph, 2015; Canadian Centre for DNA Barcoding www.ccdb.ca); qPCR can be accomplished for as little as $5/2hrs for 96 specimens using a high- throughput machine like the StepOnePlus Real Time PCR system, once an assay has been developed. Real-time PCR utilizing this assay has a wide breadth of application in larval ecology including fragmentary analysis (Jackson et al. 2012), gut content analysis

(Jarman et al. 2004), and plankton enumeration and identification (Coyne et al. 2005).

This real-time assay can be effectively and efficiently applied to major environmental assessments of industrial effects on ichthyoplankton, including the entrainment of fish larvae associated with cooling water intake structures at power- generating and manufacturing facilities. For example, a major hindrance to accurately estimating entrainment rates at power plants is the identification of species composition

(The Committee on Entrainment 1978). Visual identification of entrainment samples remains a challenge because of the degradation of samples (Azila & Chong 2010; Rabin

2010), yet this information is key to assessing ecological impact. The success of the assay described in this laboratory study, and the validation of the primer and probe set on a high-throughput platform, suggests this method would be a suitable means to address the presences of a target species in entrainment ichthyoplankton samples. Future research will involve expansion of this assay to include other species of commercial and ecological importance in the Great Lakes that may be at risk of entrainment is highly

140 feasible. The use of multiple primer and probe sets in a multiplexed real-time PCR assay would simultaneously identify several species of interest, further reducing the costs of monitoring the presence of these species.

141 5.6. References

Ager D.D., Cord I., Patrick P.H. 2006. Entrainment sampling at Darlington Nuclear Generating Station – 2004. Report Number NK38-REP-07264-10001-R00

Auer N. A. (Editor). 1982. Identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage. Great Lakes Fishery Commission, Ann Arbor, MI 48105. Special Pub. 82-3: 744 pp.

April J., Mayden R.L., Hanner R.H., Bernatchez L. 2012. Genetic calibration of species diversity among North America’s freshwater fishes. PNAS 108(26): 10602- 10607.

Azila A., Chong V.C. 2010. Multispecies impingement in a tropical power plant, Straits of Malacca. Marine Environmental Research 70: 13-25

Bayha K.M., Graham W.M., Hernandez F.J., Jr. 2008. Multiplex assay to identify eggs of there fish species from the northern Gulf of Mexico, using locked nucleic acid Taqman real-time PCR probes. Aquatic Biology 4: 65-73.

Bernatchez L., Dodson J.J. 1991. Phylogeographic structure in mitochondrial DNA of the Lake Whitefish (Coregonus clupeaformis) in North America and its relationships to Pleistocene glaciations. Evolution 45(4): 1016-1035.

Brown T. 2007. Whitefish Investigations 2006 Summary. B-Rep-00531-00017. Bruce Power Report.

Brown T. 2008. Whitefsih Investigations 2007 Summary. B-Rep-00531-00022. Bruce Power Report.

Bustin S.A. 2005. Real-Time PCR. Encyclopedia of Diagnostic Genomics and Proteomics. University of London, London, U.K. 1117-1125.

Casselman J.M., Collins J.J., Crossman E.J. et al. 1981. Lake whitefish (Coregonus clupeaformis) stocks of the Ontario waters of Lake Huron. Canadian Journal of Fisheries and Aquatic Science 38: 1772-1789.

Coyne K.J., Handy S.M., Demir E. et al. 2005. Improved quantitative real-time PCR assays for enumeration of harmful algal species in field samples using an exogenous DNA reference standard. Limnology and Oceanography: Methods. 3: 381-391.

Crawford S., Muir A., McCann K. 2001. Ecological basis for recommendations of 2001 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Chippewas of Nawash First Nation 1-168.

Cucin D., Faber D.J. 1985. Early life studies of lake whitefish (Coregonus clupeaformis),

142 cisco (Coregonus artedii) and yellow perch (Perca jlavescens) in Lake Opeongo, Ontario. Ont. Fish. Tech. Rep. Ser. No. 16

Davis B.M., Todd,T.N. 1998. Competition between larval lake herring (Coregonus artedi) and lake whitefish (Coregonus clupeaformis) for zooplankton. Canadian Journal of Fisheries and Aquatic Science 55: 1140-1148.

Freyhof J., Kottelat M. 2008. Coregonus pidschian. In: IUCN 2011. IUCN Red List of Threatened Species. Version 2011.1

Gleason L.U., Burton R.S. 2011. High-throughput molecular identification of fish eggs using multiplex suspension bead arrays. Molecular Ecology Resources 12,1, 57-66.

Golder Associates Ltd. 2008. Bruce New Nuclear Power Plant Project Environmental Assessment: Methods, Project Description, and Existing Environment. Environmental Impact Statement Volume 1 September 2008.

Golder Associates Ltd. 2008. Whitefish Emergent Larval Trap and Tow Study 2006. Bruce Power. Bruce Power Limited 05-1112-070. 34pp.

Hare J.A., Cowen R.K., Zehr J.P. et al. 1994. Biological and oceanographic insights from larval labrid (Pisces: Labridae) identification using mtDNA sequences. Marine Biology 118: 17-24.

Hebert P.D.N., Cywinska A., Ball S.L., DeWaard J.R., 2003. Biological identifications through DNA barcodes. Proceedings of the Royal Society of London. 270: 313-321.

Holmes J.A., Noakes D.L.G. 2002. Whitefish Interactions with Nuclear Generating Stations. Recommendations for Monitoring the Effects of the Bruce Nuclear Development on Whitefish in Lake Huron. Report: Axelrod Institute of Ichthyology. 173 pp.

Huang K.S., Lee S.E., Yeh Y., et al. 2010. Taqman Real-Time Quantitative PCR for Identification of Western Flower Thrip (Frankliniella occidentalis) for Plant Quarantine. Biology Letters. 6: 555-557.

Hubert N., Hanner R., Holm E., et al. 2008. Identifying Canadian Freshwater Fishes through DNA Barcoding. PLoS One 3(6): 1-8.

Ivanova N.V., Zemlak T.S., Hanner R.H., Hebert P.D.N. 2007. Universal primer cocktails for fish DNA barcoding. Molecular Ecology Notes. 7:544-548.

Jackson C.B., Gallati S., Schaller A. 2012. qPCR-based mitochondrial DNA quantification: Influence of template DNA fragmemtation on accuracy. Biochemical and Biophysical Research Communications. 423: 441-447.

143 Jarman S.N., Deagle B.E., Gales N.J. 2004. Group-specific polymerase chain reaction for DNA-based analysis of species diversity and identity in dietary samples. Molecular Ecology 13:1313-1322.

Ko H.L., Wang Y.T., Chiu T.S. et al. 2013. Evaluating the Accuracy of Morphological Identification of Larval Fishes by Applying DNA Barcoding. PLoS One. 8(1): 1-7.

Kochzius M., Seidel C., Antoniou A. et al. 2010. Identifying Fishes Through DNA Barcodes and Microarrays. PLoS One. 5(9):1-15.

LaRiviere C.M., Crawford S.S. 2013. Indigenous principles of wild harvest and management: an Ojibway community as a case study. Human Ecology. 41: 947-960.

Lu G., Basley D.J., Bernatchez L. 2001. Contrasting patterns of mitochondrial DNA and microsatellite introgressive hybridization between lineages of lake whitefish (Coregonus clupeaformis); relevance for speciation. Molecular Ecology. 10: 965-985.

Mackie I.M., Pryde S.E., Gonzales-Sotelo C. et al. 1999. Challenges in the identification of species of canned fish. Trends in Food Science and Technology 10(1): 9-14.

Mcbeath A.J.A., Penston M.J., Snow M. et al. 2006. Development and application of real-time PCR for specific detection of Lepeophtheirus salmonis and Caligus elongates larvae in Scottish plankton samples. Diseases of Aquatic Organisms. 73:141-150.

Morh L.C., Nalepa T.F. (Editors) 2005. Proceedings of a workshop on the dynamics of lake whitefish (Coregonus clupeaformis) and the amphipod Diporeia spp. in the Great Lakes. Great Lakes Fishery Commission Technical Report 66.

Naaum A.M., Foottit R.G., Maw H.E.L., Hanner R. 2012. Differentiation Between Aphis pomi and Aphis Spiraecola Using Multiplex Real-Time PCR Based on DNA Barcode Sequences. Journal of Applied Entomology. 136(9):704-710.

Pan M,. McBeath A.J.A., Hay S.J. et al. 2008. Real-time PCR assay for detection and relative quantification of Liocarcinus depurator larvae from plankton samples. Marine Biology. 153:859-870.

Pegg G.G., Sinclair B., Briskley L., Aspden W.J. 2006. MtDNA barcode identification of fish larvae in the southern Great Barrier Reef-Australia. Scientia Marina 70(S2): 7-12.

Pfrender M.E., Hawkins C.P., Bagley M. et al. 2010. Assessing Macroinvertebrate biodiversity in freshwater ecosystems: Advances and Challenges in DNA-based approaches. Quarterly Review of Biology. 85: 319-340.

Quinteiro J.C., Sotelo G., Rehbein H. et al. 1998. Use of mtDNA Direct Polymerase Chain Reaction (PCR) Sequencing and PCR-Restriction Fragment Length Polymorphism

144 Methodologies in Species Identification of Canned Tuna. Journal of Agriculture and Food Chemistry. 46(4):1662-1669.

Rabin K. 2010. Reeling in New York’s Aging Power Plants: The Case for Fish- Friendlier Power. Network for New Energy Choices. New York. 24p.

Rasmussen Hellberg R.S., Morrissey M.T., Hanner R.H. 2010. A Multiplex PCR Method for the Identification of Commercially Important Salmon and Trout Species (Oncorhynchus and Salmo) in North America. Journal of Food Science. 75(7):C595- C606.

Rasmussen Hellberg R.S., Morrissey M.T. 2011. Advances in DNA-Based techniques for the detection of seafood species substitution on the commercial market. Journal of Laboratory Automation. 16: 308-322.

Ratnasingham S., Hebert P.D.N. 2007. BOLD: The Barcode of Life Data System (http://www.barcodinglife.org). Molecular Ecology Notes. 7(3):355-364.

Roth B.M., et al. 2012. Fishes and Decapod Crustaceans of the Great Lakes Basin. In: Great Lakes Policy and Management: A Binational Perspective. Eds. Taylor, W.W. and A. Lynch. Michigan State University Press.

Schlei O.L., Crete-Lafreniere A., Whiteley A.R. et al. 2008. DNA barcoding of eight North American coregonine species. Molecular Ecology Resources. 8(6):1212-1218.

Scott W.B., Crossman E.J. 1973. Freshwater Fishes of Canada. Fisheries Research Board of Canada, Ottawa. 966 pp.

Singer G.A.C., Hajibabaei M. 2009. iBarcode.org: web-based molecular biodiversity analysis. BMC Bioinformatics. 10(Suppl 6): S14.

Tamura K., Dudley J., Nei M., Kumar S. 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution. 24: 1596-1599.

Taylor M.I., Fox C., Rico I., Rico C. 2002. Species-specific TaqMan probes for simultaneous identification of (Gadus morhua L.), haddock (Melanogrammus aeglefinus L.) and whiting (Merlangius merlangus L.) Molecular Ecology Notes. 2:599-601.

Teletchea F. 2009. Molecular identification methods of fish species: reassessment and possible applications. Rev. Fish. Biol. Fisheries. 19: 265-293.

The Committee on Entrainment. 1978. Introduction. Pages 1-18 in Schubel, J. R. and B. C. Marcy, Jr. (Editors) Power Plant Entrainment: A Biological Assessment. Academic Press, New York.

145 Todd T.N., Smith G. R. 1992. A review of differentiation in Great Lakes ciscoes. In Biology and management of coregonid fishes. Edited by T.N. Todd and M. Luczynski. Polski Arch Iwuh Hydrobiologii. 39: 13-19.

Todd T.N., Stedman R.M. 1989. Hybridization of ciscoes (Coregonus spp.) in Lake Huron. Canadian Journal of Zoology. 67:1679-1685.

Ward R.D., Hanner R., Hebert P.D.N. 2009. The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology 74: 329-356.

Valdez-Moreno M., Vasquez-Yeomans L., Elias-Gutierrez M., Ivanova N.V., Hebert P.D.N. 2010. Using DNA barcodes to connect adults and early life stages of marine fishes from the Yucatan Peninsula, Mexico: potential in fisheries management. Marine and Freshwater Research 61(6): 655-671.

Victor B. C., Hanner R., Shivji M., Hyde, J., Caldow, C. 2009. Identification of the larval and juvenile stages of the Cubera Snapper, Lutjanus cyanopterus, using DNA barcoding. Zootaxa 2215: 24-36.

Ward R.D., Hanner R., Hebert P.D.N. 2009. The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology 74: 329-356.

Ye L., Coulouris G., Zaretskaya I., Cutcutache I., Rozen S., Madden T. L. 2012. Primer- BLAST: A tool to design target-specific primers for polymerase chain raction. BMC Bioinformatics 13:134.

146

4.5

4 y = -0.2805x + 8.7548 3.5 R² = 0.99834

3

2.5

2

1.5

1 Log DNA concentration concentration DNA Log

0.5

0

-0.5 10 20 30 40 Average FAM Ct value

Figure 5.1. Standard curve generated from 10-fold serial dilutions of Lake Whitefish (Coregonus clupeaformis ) DNA from 7.1 ng/µl to 0.71 pg/µl. FAM – fluorescent reporter 6-carboxyfluorescein.

147

Figure 5.2. Melt curve analysis peaks for: (A) no template control; (B) non-target species Longnose Sucker (Catostomus catostomus); (C) non-target species White Sucker (Catostomus commersonii); and (D) target species Lake Whitefish (Coregonus clupeaformis). Only one distinct melt peak is present in panel D for Lake Whitefish. No secondary peaks indicate no primer dimer formation.

148 A

B

Figure 5.3. Results of agarose gel electrophoresis of: (A) traditional PCR using universal fish primers for the DNA barcode region; and (B) amplification with the Coregonus- specific primers designed in this study. (A) traditional PCR for identification of Longnose Sucker (LS; Catostomus catostomus) and Lake Whitefish (LW; Coregonus clupeaformis) using universal fish primers. NTC =no-template control. Bands are present for both Longnose Sucker (lane 5) and Lake Whitefish (lane 6) at the expected size of 650 bp. No bands are present for no-template controls. (B) Results of agarose gel electrophoresis of Real time PCR product using primers designed in this study. NTC=no- template control, non-target in lanes 3-6 include 2 replicates of Longnose Sucker (lanes 3,4) and White Sucker (lanes 5, 6). Bands observed for Lake Whitefish (LW) in lanes 7- 10 at approximately 120 bps, which is expected length of product amplicon from Real time PCR reaction. No bands present for the no-template control and non-target species.

149 Table 5.1. Sequences of species-specific primer and probe set for Coregonus clupeaformis. The amplicon length for the primers is 122bp of the DNA barcode region. All primers and probes target the cytochrome C oxidase subunit I (COI) DNA barcode region. FAM – fluorescent reporter 6-carboxyfluorescein. BHQ – black hole quencher 1.

Primer name Primer sequence (5′-3′) Coregonus clupeaformis Forward GCC CTA AGC CTT TTA ATC Coregonus clupeaformis Reverse GGC ATA ACT ATA AAG AAA ATC ATA Coregonus clupeaformis Probe 6FAM-CC GTG ACG ATC ACA TTA TAA ATC TGA T-BHQ1

150 Table 5.2. Species data for sequences used in the design of the primer/probe set. Barcode of Life Data System (BOLD; http://boldsystems.org) project code indicates the projects that contained sequences used with the number of sequences taken from each project in brackets; the number of sequences for each species and haplotypes per species are also indicated. Full specimen details for sequences are available from the BOLD Dataset ‘Dataset of Coregonids’, project code DS-CORG.

Species BOLD project code Number of Number of sequences haplotypes Coregonus artedi BCF (7), GBGCA (2), 23 5 SBOLW (14) Coregonus autumnalis ANGBF(1), BCF(6), 9 3 GBGCA(2) Coregonus clupeaformis BCF(8), CYTC(6), GBGCA(2), 135 15 LWLHW(66), SBOLW(27), SLHL(26) Coregonus hoyi BCF(5), GBGCA(2), SLHL(31) 38 6 Coregonus huntsman BCF(1), GBGCA (1) 2 2 Coregonus kiyi BCF(1), GBGCA (1) 2 2 Coregonus laurettae ANGBF(1), BCF(7), 9 5 GBGCA (1) Coregonus maraena IFCZE 6 2 Coregonus nasus ANGBF(1), BCF(7), 12 4 GBGCA(4) Coregonus nigripinnis BCF(2), GBGCA(1) 3 3 Coregonus peled GBGCA(1), IFCZE(6) 7 7 Coregonus pidschian ANGBF(1), GBGCA(1) 2 1 Coregonus sardinella ANGBF(1),BCF(6), 9 4 GBGCA(2) Coregonus zenithicus BCF(3),GBGCA(1) 4 2 Prosopium coulterii BCF(2), GBGCA(2) 4 4 Prosopium cylindraceum BCF(9), GBGCA(4), 21 9 LWLHW(8) Prosopium williamsoni BCF(11), GBGCA(5) 16 7 Stenodus leucichthys ANGBF(1), BCF(8), 12 3 GBGCA(3)

151 Table 5.3. Location of forward and reverse primer on the 650 base pair amplicon of the DNA barcode region of cytochrome c oxidae subunit I (COI) and corresponding probe location and alignment among Coregonus species. Created by using the NCBI-Primer BLAST making tool. **primers do not bind to the genus Prosopium, there is one base pair difference present.

Species Forward Probe (5’-3’) Probe Reverse Primer 6FAM-CC GTG ACG ATC ACA TTA TAA ATC TGA T-BHQ1 Location

C. clupeaformis 40-57 ...... 98-124 138-161 C. pidschian 40-57 ...... 98-124 138-161 C. ussuriensis 25-42 ...... T ...... 83-109 123-146 C. albula 43-60 ...... T ...... 101-127 141-164 C. peled 37-54 ...... T ...... 95-121 135-158 C. maraena 37-54 ...... T ...... 95-121 135-158 C. lavaretus 51-68 ...... T ...... 109-135 149-172 C. zenithicus 40-57 ...... T ...... 98-124 138-161 C. sardinella 40-57 ...... T ...... 98-124 138-161 C. pollan 40-57 ...... T ...... 98-124 138-161 C. nasus 40-57 ...... T ...... 98-124 138-161 C. muksun 40-57 ...... T ...... 98-124 138-161 C. migratorius 33-50 ...... T ...... 91-117 131-154 C. kiyi 40-57 ...... T ...... 98-124 138-161 C. laurettae 40-57 ...... T ...... 98-124 138-161 C. huntsmani 40-57 ...... T ...... 98-124 138-161 C. hoyi 40-57 ...... T ...... 98-124 138-161 C. autumnalis 40-57 ...... T ...... 98-124 138-161 C. artedi 40-57 ...... T ...... 98-124 138-161 C. nigripinnis 41-61 ...... T ...... 102-128 142-165 S. leucichthys ** ...... T ...... 138-161 142-165 P. cylindraceum ** ...... T ...... T ... . 98-124 **

152 P. coulterii ** .T ..A ...... 98-124 ** P. williamsoni ** .. ..A ... ..T ...... T ... . 98-124 ** P. spilonotus ** .. ..A ... ..T ...... T ... . 98-124 ** P. gemmifer ** .. ..A ... ..T ...... T ... . 98-124 ** P. abyssicola ** .. ..A ... ..T ...... T ... . 98-124 **

153 Table 5.4. DNA mixtures containing different proportions of Coregonus clupeaformis (CC) [7.1 ng/ µl] mixed with non-target species Coregonus hoyi (CH) [7.1 ng/ µl] and Sander vitreus (SV) [7.1 ng/ µl]. Absolute concentration of CC DNA in each mixture provided. Ct values for each replicate shown along with average Ct value.

Proportion CC (µl) SV (µl) CH (µl) Ct Ct [CC] of CC average 50% 15 0 15 18.09 17.97 5.13 18.2 17.61 50% 15 15 0 18.84 18.27 4.3 18.12 17.84 0% 0 15 15 NA NA 0 NA NA 33% 10 10 10 18.51 18.44 3.4 18.52 18.72 17% 5 10 15 20.09 20.09 1.16 19.98 20.2 10% 3 12 15 20.42 20.27 1.16 20.3 20.09

154 Table 5.5. Validation of protocol and primer and probe set on StepOnePlus Real-Time PCR machine and comparison of Ct values. All samples extracted from larval caudal fin clips, except for Salmo salar and Prosopium cylindraceum, which were collected from adult muscle samples. [DNA] in ng/ µl. CC= Coregonus clupeaformis, CH= Coregonus hoyi, SV= Sander vitreus. See Table II for [CC] DNA (ng/ µl) in mixtures 1-5. DNA concentrations of samples used listed in methods for each species.

Species Ct (Cepheid) Ct (StepOnePlus) C. clupeaformis 17.49 17.81 C. clupeaformis 20.56 21.58 C. clupeaformis 21.89 22.06 C. clupeaformis 20.49 20.98 C. clupeaformis 20.68 21.23 S. salar NA NA E. exile 31.99 31.67 P. flavescens 29.66 29.89 O. mordax NA NA N. buchanani NA NA C. artedi NA NA C. hoyi NA 30.31 P. cylindraceum NA NA C. commersonii NA NA C. catostomus NA NA Mix 1 (CC:CH) 17.97 19.22 Mix 2 (CC:SV) 18.27 18.39 Mix 3 (SV:CH) NA NA Mix 4 (CC:CH:SV) 18.44 19.45 Mix 5 (CC:CH:SV) 20.09 20.29

155 Chapter 6. The effect of environmental conditions on the distribution and abundance of larval Lake Whitefish (Coregonus clupeaformis) in nearshore embayments at Douglas Point, Lake Huron.

6.1. Abstract

The goal of this study was to investigate the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish in nearshore embayments at

Douglas Point, Lake Huron. Three cause-effect hypotheses were evaluated as potential determinants of larval Lake Whitefish distribution/abundance: the Larval Pulse

Hypothesis, the Longshore Transport Hypothesis, and the Embayment Nursery

Hypothesis. Surface plankton tows and environmental data were collected from ice-out to early summer during 2014. Plankton tows revealed relatively high densities of larval

Lake Whitefish in nearshore waters of Inverhuron Bay, compared to the other nearshore embayments at Douglas Point and previous ichthyoplankton studies in the region.

Overall, there was little relationship between larval Lake Whitefish densities and environmental variables (i.e. temperature, pH, conductivity, dissolved oxygen).

Redundancy Analysis revealed that week and wind direction were significant factors explaining larval Lake Whitefish distribution and abundance. In particular, week 2 of sampling (mid-May) and northeasterly (onshore) winds were identified as the most important conditions of these factors for explanation of ichthyoplankton densities. Based on the results of this investigation, there was strong support for the Larval Transport

Hypothesis, little support for the Longshore Transport Hypothesis, and equivocal support for the Embayment Nursery Hypothesis – however, the latter might become more important when information becomes available about subsurface distribution/abundance and potential schooling behaviour by larval Lake Whitefish.

156 6.2. Introduction

In the Laurentian Great Lakes, Lake Whitefish (Coregonus clupeaformis) typically spawn in the autumn months on nearshore rocky substrates (Ebener et al. 2008;

Lynch et al. 2015), with embryos overwintering on those spawning grounds until early spring when they hatch as negatively-buoyant free embryos and swim up into the water column and associated currents in search of exogenous food (Balesic & Martin 1987;

McKenna & Johnson 2009; Pothoven et al. 2014). Individual Lake Whitefish free embryos have a critical period of approximately 14 days during which they must switch to exogenous food before exhaustion of endogenous energy reserves and death by starvation (O’Connell & Raymond 1970; Brown & Taylor 1992; Claramunt et al. 2010a).

Upon first exogenous feeding, Lake Whitefish larvae must continue to actively pursue and prey upon zooplankton that can exhibit major daily vertical migrations in the water column (Taylor & Freeberg 1984; Freeberg et al. 1990).

Despite decades of research, the early life history and ecology of larval Lake

Whitefish remains poorly understood (Hoagman 1973; Cucin & Faber 1985; Claramunt et al. 2010a). Pritt et al. (2015) suggest that understanding the ecology of larval fishes is paramount in understanding spawning activity of fishes and increasing long-term monitoring and habitat restoration. Several authors have hypothesized that a variety of local environmental and ecological conditions can play major roles in determining survival rates of Lake Whitefish during the larval period, including: wind-induced surface currents (Brown et al. 1993; McKenna & Johnson 2009; Lynch et al. 2015), basin and regional geostrophic water currents (Clady & Hutchinson 1975; Taylor et al. 1987), proximity of nearshore embayments with protected waters (McKenna & Johnson 2009),

157 temperature and/or dissolved oxygen (Lane et al. 1996; Ryan & Crawford 2014; Lynch et al. 2015), ice and light availability (Freeberg et al. 1990; Hoyle 2003; Roseman et al.

2012), regional nutrient loading (Mills 1985), and rates of phytoplankton and invertebrate zooplankton productivity (Taylor & Freeberg 1984; Freeberg et al. 1990; Lane et al.

1996).

Douglas Point, Lake Huron has been the focus of several Lake Whitefish ecological studies focusing on Environmental Assessment (EA) of the effects associated with operation of the Bruce Nuclear Generating Station (BNGS), North America’s largest nuclear power station (Figure 1.1). The BNGS complex is comprised of two distinct facilities (Bruce A Units 1-4, Bruce B Units 5-8), each of which has its own once-through cooling water system that draws Lake Huron water from intakes 800-1100 m offshore at water depths of approximately 15m (Holmes & Noakes 2002). Each facility pumps Lake

Huron intake water to a common forebay channel, from which four nuclear reactors individually pump water through their own condenser systems, which in turn cool internal reactor-generated steam before releasing the heated effluent back into Lake

Huron via a common discharge channel (ESG& BEAK 2000; Holmes & Noakes 2002).

At full operational capacity, the combined BNGS cooling water system intakes and outputs approximately 360 m3 s-1 of Lake Huron water heated to 11+C° above ambient conditions; equivalent to 10x peak spring flow of the Saugeen River, one of the largest tributaries to Lake Huron (Frank 1981; OPG 1999). The environmental and ecological effects of BNGS on Lake Whitefish have come under intense scrutiny over the past decade, largely due to the direct involvement of the Saugeen Ojibway Nation

(collectively the Chippewas of Nawash Unceded First Nation and Saugeen First Nation).

158 These First Nations have pressed for recognition of this species as a ‘Valued Ecosystem

Component’ in the federal EA process, especially due to the species’ cultural, social and economic importance for their communities (Crawford et al. 2001). Currently, the

Saugeen Ojibway Nation deploys the largest Aboriginal commercial fishery on the Great

Lakes, and the Main Basin Lake Whitefish population(s) support the bulk of their harvest

(Crawford et al. 2001). In 2003 the Canadian Nuclear Safety Commission required the development and implementation of a Lake Whitefish Follow-Up Monitoring Program to address the EA concerns of the First Nations (CNSC 2003). In 2009 the Saugeen

Ojibway First Nation and Bruce Power Limited (current lessee and operator of the BNGS facilities) entered into a bi-lateral collaborative research agreement to investigate the ecological effects of BNGS on Lake Whitefish at Douglas Point, including the distribution and abundance of larvae (Crawford et al. 2014).

There are at least five potential effects of the BNGS cooling water system on distribution and abundance of the Lake Whitefish larvae at Douglas Point: (a) intake entrainment of larvae along with Lake Huron source water at Bruce A and B intakes, (b) mechanical/thermal/contaminant mortality of entrained larvae as they are transported through the facilities, (c) output discharge of entrained larvae that have passed through the facilities, (d) output entrainment of larvae in Lake Huron receiving waters that are displaced offshore by the discharge currents, and (e) mechanical/thermal/contaminant mortality of larvae in Lake Huron receiving waters. Taken together, these potential

BNGS cause-effect mechanisms can have a profound influence on the natural distribution and abundance of Lake Whitefish larvae at Douglas Point (Wismer et al. 1986; Holmes &

Noakes 2002, Holmes et al. 2002).

159 The Douglas Point region of southeastern Lake Huron contains numerous rocky points and shoals that serve as spawning habitat for Lake Whitefish, including McCrae and Gunn Points to the south, all of the Douglas Point shoreline, and the Loscombe,

Welsh and Scougall Banks and associated shoreline to the north (Goodyear 1982: Holmes

& Noakes 2002). In terms of adjacent embayments that could serve as productive larval nursery habitat, there are Inverhuron and Holmes Bays to the south, and Baie du Doré to the north of Douglas Point. Over the past 50 years, there has been limited research/EA sampling for larval Lake Whitefish at Douglas Point (Table 1.1). The majority of this sampling effort has been deployed in offshore waters, rather than the more protected embayments of Inverhuron Bay, Holmes Bay or Baie du Doré. However, numerous investigations on early life history of Lake Whitefish elsewhere in Lake Huron and the other Great Lakes have emphasized the importance of protected embayments in providing larval nursery habitat for this species (Loftus 1978a, 1978b; McKenna &

Johnson 2009; Claramunt et al. 2010b).

The key ecological uncertainty (KEU) of this investigation is to determine the extent to which environmental conditions affect the distribution and abundance of larval

Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron. There are at least three qualitatively different cause-effect mechanisms that could be involved in determining larval Lake Whitefish distribution and abundance. The Larval Pulse

Hypothesis states that spring hatching of Lake Whitefish and corresponding peaks in spring primary/secondary biomass (zooplankton prey) typically associated with nearshore

Great Lakes waters after ice-out, lead to a distinct pulse in larval abundance that reduces as ichthyoplankton deplete their prey resources (Faber 1970; Bidgood 1974, Peeters et al.

160 2007). The Longshore Transport Hypothesis states that ichthyoplankton in general, and larval Lake Whitefish in particular, are consistently moved by the prevailing geostrophic longshore currents (Freeberg et al. 1990; Claramunt et al. 2010a), which in the Douglas

Point region of Lake Huron flow in a northerly direction 80% of the time (Wismer et al.

1986). The Embayment Nursery Hypothesis states that nearshore embayment morphology causes eddies in water currents that retain both ichthyoplankton (including

Lake Whitefish larvae) and their zooplankton prey (Zhao et al. 2012; Howell et al. 2014;

Kikko et al. 2015).

The goal of this investigation is to determine the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron. In order to achieve this goal, the following objectives must be satisfied: (1) collect ichthyoplankton samples and associated environmental data according to a spatio-temporally stratified sampling design at

Douglas Point; (2) determine which space-time variables have a major effect on environmental conditions; (3) determine which space-time variables have a major effect on the distribution and abundance of larval Lake Whitefish at Douglas Point; and (4) determine which space-time variables/environmental conditions have a major effect on the distribution and abundance of larval Lake Whitefish at Douglas Point.

6.3. Material and Methods

Study Design

The 2014 sampling program focused on three nearshore embayments adjacent to

Douglas Point, Lake Huron: Inverhuron Bay and Holmes Bay south of the BNGS B

161 Intake and Baie du Doré north of the BNGS A discharge (Figure 1.1). For the

Inverhuron/Holmes region (Figure 6.1A), there were eight within-embayment plankton- tow transects, six of which followed the 2m bathymetric shoreline contour, and two of which were located in the middle of the embayment. The Inverhuron/Holmes region also had five transects located outside the embayments: one nearshore transect south of the embayments, two deeper transects adjacent to Inverhuron Bay, one deeper transect adjacent to Holmes Bay, and one nearshore transect north of Holmes Bay just south of

Douglas Point. There were a total of nine stations in Inverhuron/Holmes that were used for water pump sampling of nearshore waters, as indicated by triangles in Figure 6.1A.

For the Baie du Doré region (Figure 6.1B), bathymetric complexity and shallow water allowed four within-embayment plankton-tow transects, two of which were conducted in relatively small 2-3m deep water pockets on the south side of the embayment, while the other two transects followed the 3m bathymetric shoreline contour on the north side of the embayment. The Baie du Doré region also had three transects located outside the embayment: two deeper transects adjacent to the embayment and one nearshore transect north of the embayment. There were a total of seven stations in Baie du Doré that were used for water pump sampling of nearshore waters, indicated as triangles in Figure 6.1B.

Thus, there were a total of 20 transects for plankton tows and 16 water pump stations established in association with the three embayments: Inverhuron Bay=8/8, Holmes

Bay=5/1, Baie du Doré=7/7 (Table 6.1).

RainWise WindLog wind data loggers (RainWise Inc., Trenton, ME, USA) were deployed on 2 May 2014 at Gunn Point, Inverhuron (N 44.29775, W 081.60471) and at

Scotts Point, Baie du Doré (N 44.33551, W 081.55540) facing north. The wind loggers

162 were programmed for 15-minute recordings of average wind speed (km h-1) and average wind direction (° departure from North), which was later converted to a code representing wind destination according to the 8 cardinal compass directions (e.g., N, NE, E, etc.). For statistical analyses, each sample day was associated with average wind speed and modal wind direction for the time period 04:00-12:00. For the purposes of this study, wind direction will here by be expressed in terms of destination rather than origin.

Hobo TidbiT v2 temperature loggers (Onset Hobo® Data Loggers, Bourne, MA,

USA) were programmed to record every 15 minutes, affixed to cinder blocks, and deployed at the center of each 2m contour transect in Inverhuron/Holmes Bays and Baie du Doré on 05 and 11 May 2014, respectively (Figure 6.1A-B, represented by asterisks).

Environmental and plankton sampling occurred during a six-week period extending from shortly after ice-out on 5 May to 21 June 2014 (Table 6.2). Sequence of weekly sampling events was determined foremost by safety in a 5m open, aluminum boat, then by wind/wave conditions and accessibility of transects/stations with prioritization as follows: Baie du Doré>Holmes Bay>Inverhuron Bay; pump sampling>transect sampling, outer embayment>inner embayment.

Water conditions were assessed weekly at the midpoint of each transect, using a

YSI EXO2 Sonde (Yellow Springs, Inc., Yellow Springs, Ohio, USA): depth (m), temperature (°C), specific conductivity (µs cm-1), salinity (psu), dissolved oxygen (mg L-

1), turbidity (FNU) and pH. The sonde was calibrated following manufacturer guidelines at the beginning of sampling and repeated every two weeks through the sampling period.

For deployment, the sonde was lowered to 1m depth for 60s equilibration, and then data were recorded for an additional 60s. Probe malfunction occurred during part of Week 3,

163 leading to partial data being collected for samples collected during that time period.

Ichthyoplankton samples were obtained from horizontal surface tows of a submerged 500µm mesh plankton net (0.75m hoop diameter, 2.3m length; Halltech

Environmental Inc., Guelph, Ontario, Canada) using a 5m Bayview aluminum boat with an Envinrude ETEC 25hp outboard motor; tows were oriented into the prevailing wind, where possible. A calibrated flow meter (General Oceanics Model 2030R) was installed in the mouth of the net, and the meter data were recorded at the beginning and end of each tow to allow for calculation of water volume sampled, according to instructions provided by the manufacturer. At the end of each plankton tow, the contents of the cod end were washed into a sealed plastic container until processed on shore, then transferred into small glass sampling containers with 95% ethanol, which was refreshed in the containers after a few days to remove water and ensure proper preservation. A Honda

WB20Xt2C 50mm water pump was also used to collect plankton samples at 16 stations in nearshore water less than 1m depth, where the plankton nets could not be deployed

(triangles in Figures 6.1A-B). Pump flow rates were calibrated prior to each deployment and then operated for 15 minutes per station in combination with a floating 500µm mesh plankton net basket tied to the boat (30x48x60cm deep). At the end of Week 3, pump sampling had not yielded any ichthyoplankton, and this form of sampling effort was discontinued. The number of fish larvae caught in each sampling event was converted to a standard density expressed as # individuals 1000m-3, which is consistent with other larval Lake Whitefish assessments in the Great Lakes basin (e.g., McKinley & Sim 1986,

Ryan & Crawford 2014).

164 Species Identification

All larval fish were visually identified to species using a dissecting microscope

(Olympus SZ-CTV) and the Auer (1982) identification key, which uses melanophore patterns as a key identifying feature to distinguish larval Lake Whitefish from other species. It should be noted that Auer (1982) specifically states that:

“This manual for the identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage treats 24 families and 145 species. In addition to a key to the families, keys were constructed for selected species within some families...A description of the egg is given, as are morphometric, morphological and pigmentation characteristics of yolk-sac larvae, larvae and juveniles” (Auer 1982 p. X)

However at the time of this study there were no other species key available for Great

Lakes ichthyoplankton.

After visual species identification, a caudal fin clip was obtained from each specimen for DNA extraction using Qiagen DNeasy Blood and Tissue Kits (Qiagen,

Valencia, CA) following the manufacturer instructions. PCR amplification was carried out in 12.5 µl reaction volumes with 6.25 µl 10% trehalose, 2 µl ddH2O, 1.25 µl 10X buffer, 0.625 µl MgCl2 (50 mM), 0.1 µl each of forward and reverse 10 µM concentration

PCR primers (universal fish primers VF1i_t1 and VR1i_t1; Ivanova et al. 2007), 0.0625

µl 10 mM dNTPs, 0.06 µl Platinum Taq polymerase (5 U µl-1), and 2 µl of DNA. The reaction profile was: an initial hot start at 94°C for 120sec; followed by forty cycles with a denaturation at 94°C for 30 sec, annealing temperature of 52°C for 40 sec, and extension at 72°C for 60sec; followed by a final extension at 72°C for 10min and hold at

4°C indefinitely. Amplification success of PCR products was ascertained visually using

165 pre-cast 2% Agarose E-gels (Invitrogen), where a clear, unambiguous band on the gel indicated successful amplification.

DNA sequencing reactions were carried out in 14 µl reaction volumes with 1 µl

BigDye v3.1, 1 µl 5X SeqBuffer, 1 µl primer (10 µM; C_VF1LFt1 for forward and

C_VR1LRt1 for reverse), 10 µl ddH2O, and 1.5 µl PCR product. PCR products were bidirectionally sequenced using an ABI 3730 DNA Analyzer (Applied Biosystems). Bi- directional sequence contig assemblies were created and edited using Sequencher v.4.9

(Gene Codes), and multiple sequence alignments were generated manually using BioEdit v.7.0.5.3 (Hall 1999). Sequences were uploaded to the Barcode of Life Data System

(BOLD; Ratnasingham & Hebert 2007) public project titled “Douglas Point, Lake Huron

Larval Fish” (project code: DPLHF) and subsequently submitted to GenBank (Accession

Numbers KR733317-KR733425) via BOLD. All barcodes were queried against the sequences of known provenance using the BOLD ID engine (www.boldsystems.org,

Ward et al. 2009) using the “species ID” option. A DNA barcode species match was defined as a percent of sequence similarity to a unique species in the reference library greater than 99%.

All Real-Time PCR (qPCR) reactions were carried out in 25 µl volumes containing 17 µl ddH20, 5 µl template DNA, 1.25 µl forward primer [0.5 µM], 1 µl reverse primer [0.4 µM], and 0.75 µl probe [0.3 µM]. OmniMix HS mastermix

(Cepheid) was used according to manufacturer guidelines. PCR cycling conditions were an initial hold step of 95oC for 120 seconds, followed by 35 cycles of 95oC for 11 seconds, 62oC for 30 seconds, and 72oC for 10 seconds on the StepOnePlus Real- Time

PCR system (Applied Biosystems). All reactions were carried out using MicroAmp fast

166 96-well reaction plates (100 µl) (Applied Biosystems). The fluorescence threshold was manually set to 8000 units to ensure this reading was in the middle of the amplication portion of the threshold cycle (Ct) curve. Two replicates of each sample were tested along with no-template control wells on each plate. Reference thresholds of Ct values can be used to indicate relative measure of target species in a sample, where: Ct<30 indicates occurrence of target Lake Whitefish, Ct≥30 indicates possible occurrence of target; Ct=NA ‘not applicable’ indicates absence of target.

Statistical Analyses

The statistical distribution of environmental variables was examined using boxplots and frequency histograms to visually assess normality and investigate any outliers. Most variables appeared reasonably normal from histogram distribution plots; pH did not need to be transformed because it was already on a log scale (Nabout et al.

2009). Temperature was non-normal in distribution, and these data were transformed using a log(x+1) conversion, based on recommendations by Gasiunaite et al. (2005) and

Peng et al. (2012).

Data were imported into CANOCO v.5.0 (Microcomputer Power, Ithaca, USA; ter Braak & Smilauer 2012) into three tables; (a) spatial/temporal data as factors (i.e. week, bay, embayment: in/out, offshore: near/far); (b) environmental conditions (i.e. wind direction, wind speed, log(x+1)temperature, dissolved oxygen, pH); (c) standard density of Lake Whitefish and ‘Other Ichthyoplankton.’ Redundancy Analysis (RDA) was performed with forward selection of explanatory variables (Legendre &Legendre

2012; Xuemei et al. 2012) for two specific tests: effect of space-time variables on

167 environmental conditions, and effect of space-time variables/environmental conditions on ichthyoplankton standard density. See Table 6.3 for details on variables and associated codes used in the statistical analyses and associated tables and figures. For the purpose of this investigation, statistical significance was established as p<0.05.

6.4. Results

Species Identification

Table 6.4 shows the consistency between genetic (qPCR, DNA barcoding) and visual species identification for the 114 larval fish that were collected at Douglas Point embayments in 2014. DNA barcoding yielded high quality sequences for 113/114

(99.1%) of the larvae in this collection: 43 (37.7%) as Lake Whitefish (Coregonus clupeaformis), 1 Lake Herring (Coregonus artedi), 1 Round Whitefish (Prosopium cylindraceum), and the remainder as non-coregonines - 29 (25.4%) Longnose Sucker

(Catostomus catostomus), 38 (33.3%) White Sucker (Catostomus commersoni), and a single Common Carp (Cyprinus carpio). It should be noted that DNA barcoding was unable to identify one fish larva that was collected at Douglas Point in 2014.

One half (57/114=50%) of the larvae that were designated by qPCR with Ct as

'NA' indicating no Lake Whitefish DNA present in the sample; all of these specimens were identified by DNA barcording as non-salmonids (Catostomus catostomus,

Catostomus commersoni, Cyprinus carpio). For the 14/114 (12.2%) of larvae designated

Ct>30, two were identified as non-Lake Whitefish coregonines (Lake Herring Coregonus artedi, and Round Whitefish Prosopium cylindraceum), 11 were identified as catostomids, and one specimen did not yield high quality sequence. For the 43/114

168 (37.7%) of larvae designated Ct≤30, all were confirmed by DNA barcoding as the target

Lake Whitefish (Coregonus clupeaformis). Thus, given this study's a priori Ct=30 threshold for qPCR resolving target species identification, DNA barcoding confirmed

100% of the qPCR conclusions.

With regard to visual identification, in the family Salmonidae, 42/45 (93.3%) showed consistency with barcoding to the family level, and 33/45 (73.3%) showed consistency with barcoding to the species level. For Lake Whitefish specifically, 8/43

(18.6%) were visually identified as Lake Herring and 3/43 (6.9%) were visually identified as White Sucker. The single Round Whitefish was identified to the same species, while one Lake Herring was visually identified as a Lake Whitefish. In the family Catastomidae, 59/67 (88%) showed consistency with barcoding to the family level, with 8/67 (12%) identified visually as Lake Whitefish. At the species level, 0/29

(0%) showed consistency with barcoding for Longnose Sucker, which were all visually identified as White Sucker. All 38/38 (100%) White Sucker showed consistency with barcoding to the species level. In the family Cyprinidae, 1/1 (100%) showed consistency with barcoding to the family level, and 0/1 (0%) showed consistency to the species level with one Common Carp being visually identified as a Fathead Minnow.

Larval Distribution and Abundance

Weekly distribution and abundance of the n=43 Lake Whitefish larvae collected at Douglas Point in 2014 are shown in Table 6.5. The first week of sampling yielded seven larvae, peaking with 32 larvae in the second week (19-25 May), and then trailing off with one or two larvae per week until the final week of sampling with no larvae

169 collected at all. In terms of spatial distribution, it can be seen that Lake Whitefish were collected from all three embayment regions, with 34/43 (79.1%) of larvae collected in

Inverhuron Bay, 8/43 (18.6%) collected in Holmes Bay, and only a single larva (2.3%) collected in Baie du Doré during Week 5 (9-16 June). When considering the standard densities of larvae, 10/13 (77.0%) of transect samples with Lake Whitefish larvae exhibited less than 10 individuals 1000m-3. In contrast, there were three samples with extremely high standard densities: Week 1 nearshore waters outside Holmes Bay near the

BNGS B Intake (21.1 individuals 1000m-3), Week 3 nearshore waters inside Inverhuron

Bay (25.7 and 52.2 individuals 1000m-3). Appendix 6.1 provides similar spatio-temporal data for non-Lake Whitefish larvae collected during the 2014 sampling program.

Effect of Space-Time on Environment

Table 6.6 presents axis summary statistics for the RDA of space-time variables on environmental conditions sampled at Douglas Point in 2014. Taken together, the explanatory variables accounted for 35.0% of the total variation in the environmental data. The first two axes in the RDA were significant and combined to account for 67.8% of the adjusted explained variation of space-time variables on environmental conditions, with Axis 1 explaining 43.3% and Axis 2 explaining 24.5%. Figure 6.2 shows a biplot of

Axes 1 and 2, with space-time explanatory variables depicted as triangles and environmental response variables as vectors. Week features prominently on both axes, with weeks 4/6 loading strongly positive and week 1 loading strongly negative on Axis 1, along with high temperature/conductivity and low dissolved oxygen. Week 1 was notable for being distinctly positive on Axis 2 along with a relatively strong westerly wind. The

170 three are relatively close to the biplot origin; however, Inverhuron and Holmes Bays are more closely associated with low temperature and high dissolved oxygen, in contrast to

Baie du Doré, which is associated with high temperature and low dissolved oxygen.

Table 6.7 presents variance partitioning from RDA Forward Selection for whole factors of space-time explanatory variables, showing that week (30.0% total variation) and Bay

(9.7%) - especially Baie du Doré - were significant.

To further investigate the strong effect of week in this analysis, Figure 6.3 shows time series of data logger temperatures for (a) Inverhuron Bay and (b) Baie du Doré for the period May-September 2014. In both cases, ambient water temperatures showed similar 10-15 day oscillations during the first six weeks that correspond to ichthyoplankton sampling, with Inverhuron Bay cycling from approximately 6 to 12°C and Baie du Doré cycling from 10 to 16°C.

Effect of Space-Time/Environment on Ichthyoplankton Standard Density

Table 6.8 shows the axis summary statistics for the RDA of space- time/environmental conditions on the standard densities of larval fishes sampled at

Douglas Point, Lake Huron in 2014. Taken together, the combined explanatory factors accounted for 13.8% of the total variation in the standard density of larval fishes. Only

Axis 1 was significant, accounting for 77.15% of the adjusted explained variation in standard densities. Figure 6.4 shows a uniplot of Axis 1, with space-time/environmental explanatory variables depicted as triangles and standard density response variables as vectors. Of all the space-time/environment variables included in this analysis, only wind direction and week were important as explanations of standard densities for larval fishes.

171 Specifically, wind in the northeasterly direction and week 2 were strongly positively correlated with ichthyoplankton standard densities. Table 6.9 presents variance partitioning from RDA Forward Selection for whole factors of space-time/environmental explanatory variables, showing that wind direction (12.2% total variation) and week

(13.2%) were both significant.

To further investigate the effect of week and wind direction in this analysis,

Figure 6.5 shows a wind rose created to display the destination of wind for standard densities by sampling week for larvae of (A) Lake Whitefish and (B) all species combined. Concentric rings correspond to the standard density (# larval 1000m-3) of larval Lake Whitefish sampled at each event. Open circles and week numbers represent observed standard densities of larval Lake Whitefish sampled in 2014. In both cases, it is clear that the majority of ichthyoplankton samples were collected when the prevailing wind had been in a northerly or northeasterly direction - especially those samples during week 2, and with standard densities exceeding 30 individuals 1000m-3. It is important to realize that prevailing wind conditions that were associated with ichthyoplankton samples in the Redundancy Analysis had been arbitrarily defined by the eight hour period from

04:00-12:00 on the day of sampling. In order to obtain a more comprehensive representation of wind direction throughout the entire sampling period, a time series of conventional wind roses (Figure 6.6) was generated for all records during each of the six sampling weeks - including times when extreme conditions precluded safe boating for ichthyoplankton sampling. There are two major observations that can be made from these weekly wind roses: (a) the lack of consistent pattern in weekly wind direction through the

172 sampling period, and (b) the predominance of onshore winds (i.e., N,NE,E,SE) during week 2.

Serendipitous Samples of Ichthyoplankton

Independent of the spatio-temporal sampling design for this investigation, personal observations were made during the sampling period, regarding the presence of dense, nearshore ichthyoplankton aggregations in water depth less than 1m. Informal shoreline surveys were conducted in Inverhuron/Holmes Bays and Baie du Doré during weeks 3-5 in order to locate and sample these aggregations with a dip net. A total of 75 larvae were collected in Inverhuron Bay, all of which were in close proximity to the

Inverhuron Provincial Park boat launch, while a total of 35 larvae were collected in the

Baie du Doré boat launch. All of these supplemental larvae were identified to species using DNA barcoding and qPCR following the methods described above. All 75 of the

Inverhuron larvae were identified as Longnose Sucker; 14/35 (40.0%) of Baie du Doré larvae were identified as White Sucker and the remaining 21/35 (60.0%) of the larvae were identified as Longnose Sucker; qPCR identified all larvae as non-Lake Whitefish.

6.5. Discussion

Species Identification

Overall, DNA barcoding was able to identify 99.1% of the ichthyoplankton collected in this investigation from Douglas Point, Lake Huron. Given the effort that has been expended to create the BOLD-Barcode of Life Data Systems (Ratnasingham &

Hebert 2007), including entries for virtually all of the Great Lakes fishes (Hubert et al.

173 2008), the high success rate of species identification in this study should come as no great surprise. The single larva that could not be identified by DNA barcoding could have failed to amplify based on degraded DNA. It is important to note that qPCR target identifications (i.e. Coregonus clupeaformis) was 100% consistent with DNA barcoding identification, thus supporting the potential use of qPCR as a relatively simple, rapid and cost-effective solution for target identification in a multi-species sample (Pan et al. 2008;

Naaum et al. 2012). In the case of BNGS environmental assessments, federal regulators have required the selection of Lake Whitefish as a 'Valued Ecosystem Component,' which in turn requires targeted attention of this species in both monitoring and research (CNSC

2003, Crawford et al. 2014). Finally, the comparison of visual identification with genetic identification methods in this study continues to support the need for combining both methods to safeguard against errors (see Chapter 3).

Larval Distribution and Abundance

It was clear from the results of this investigation that Lake Whitefish larvae are non-randomly distributed in time and space at Douglas Point, Lake Huron. During the spring of 2014, larvae were collected during the very first week of sampling immediately after the completion of ice-out, suggesting that the hatching of free embryos was underway while ice was still in the embayments. This observation is consistent with results of a recent study by Ryan & Crawford (2014), sampling larval Lake Whitefish in

Stokes Bay, Lake Huron. The maximum larval abundance observed during week 2 sampling of this study is also consistent with observations from other larval Lake

Whitefish investigations, where a distinct pulse was noted in early May (Loftus 1978a,

174 1978b; Freeberg et al. 1990; Ryan & Crawford 2014). However, the inability to sample larvae prior to completion of ice-out prevents any strong conclusions about what was happening with embryonic hatching patterns prior to the first week of May. Given the importance of this 'critical period' in Lake Whitefish life history (Taylor & Freeberg

1984; Brown & Taylor 1992; Pothoven et al. 2014), it is clear that innovative sampling solutions are required to enhance our ability to find and collect larvae under such challenging spring boating/sampling conditions.

In terms of spatial distribution, there were several clear patterns exhibited by Lake

Whitefish larvae at Douglas Point during 2014. The vast majority (97.7.4%) of the 43 larvae were collected south of Douglas Point/BNGS, in the region of Inverhuron and

Holmes Bays, while only a single Lake Whitefish larva was collected north of that point in Baie du Doré. There are several possible factors associated with Baie du Doré that could account for the lack of Lake Whitefish larvae observed in 2014, including but not limited to: lack of internal or adjacent spawning grounds/activity, lack of appropriate currents to transport free embryos or larvae into the bay, inhospitable environmental conditions for survival (e.g., temperature, dissolved oxygen), lack of appropriate zooplankton prey, excessive predation. More detailed investigation would need to be done to help clarify which of these hypotheses contribute to explaining the lack of Lake

Whitefish larvae in Baie du Doré.

With regard to spatial distribution of Lake Whitefish larvae collected south of

Douglas Point, the fish were found in and around both embayments, sometimes inside and sometimes outside the bay proper; when inside the bay, sometimes they were found over nearshore 2m water, but sometimes further offshore. There were a couple of major

175 observations regarding the three high standard densities of Lake Whitefish observed in this study. First, the highest standard densities of larval Lake Whitefish observed in this study (52.5 individuals 1000m-3) are relatively low compared to other larval Lake

Whitefish studies in the Great Lakes, but are higher than other historically recorded larval

Lake Whitefish sampling at Douglas Point (Table 1.1 and references therein). Claramunt et al. (2010b) reported a standard density of 2,000 individuals 1000m-3, second only to

Ryan & Crawford (2014) who reported the highest larval Lake Whitefish standard density from the Great Lakes region at 2,026 individuals 1000m-31. The highest standard density reported in this study was 53 individuals 1000m-3. Second, all three high densities occurred over 2m deep nearshore waters, with two instances inside an embayment

(Inverhuron Bay) and one instance outside an embayment - between Inverhuron/Holmes and Douglas Point, immediately inshore from the BNGS B water intake (Fig 6.1A-B).

Despite the fact that this study was constrained to sampling only surface and/or nearshore waters at Douglas Point, these observations combine to raise the possibility that the observed high standard densities resulted from the horizontal surface plankton tows encountering schools of Lake Whitefish larvae. Schooling behaviour has been suggested elsewhere for larvae and juveniles of this species (Bajkov 1930; Hart 1930) and will be considered in more detail with reference to the serendipitous sampling of aggregated ichthyoplankton (see subsection below).

1 It is important to note that these numbers may be inflated because species identifications have not been confirmed by an independent technique (DNA barcoding) and could potentially contain mixed species as seen in the work in this chapter

176 Effect of Space-Time on Environment

Forward Selection of the Redundancy Analysis revealed that week and bay were the most significant space-time factors explaining variation of environmental conditions at Douglas Point, Lake Huron in 2014 (Tables 6.6, 6.7). Week 1 was noted as being extremely cold, relative to the remaining weeks - which is not surprising, given that sampling began immediately after ice-out. The fact that specific weeks (i.e., 4 and 6) correlated strongly with higher temperatures and lower dissolved oxygen levels (Figure

6.2) may be explained by the thermal oscillations observed from the water temperature data loggers throughout the sampling season (Figure 6.3 - first third of the time series). It is not clear what underlying mechanism(s) could drive these thermal oscillations; however, the closely matching pattern between Inverhuron Bay and Baie du Doré suggests a broader regional phenomenon, such as inshore/offshore winds or basin currents moving different thermal water masses through the embayments (Griffiths 1974;

Saylor & Miller 1976). Perhaps the most likely explanation for the observed thermal phase shift of 4C° ΔT at Baie du Doré relative to Inverhuron Bay is simply the chronic effect of the BNGS A thermal discharge which exists immediately at the western edge of

Baie du Doré (Figure 6.1A-B, OPG 1999; Holmes & Noakes 2002).

Effect of Space-Time/Environment on Ichthyoplankton Standard Density

Redundancy Analysis indicated that one space-time factor (week) and one environmental factor (wind direction) were significant contributors toward explaining the observed variation in ichthyoplankton standard density (Tables 6.8, 6.9). Week 2 and northeasterly winds were specifically identified as having strong positive relationships to

177 standard density (Figure 6.4), indicating that there was something important about the effect of local winds during week 2 that was particularly conducive to catching Lake

Whitefish larvae in horizontal surface plankton tows. This observation is consistent with historical and current larval Lake Whitefish sampling (e.g. Brown et al. 1993; Ryan &

Crawford 2014). The fact that this environmental relationship existed for larvae of both

Lake wWhitefish, and the other five species sampled at Douglas Point during 2014, suggests that wind direction may be a generally important condition affecting ichthyoplankton distribution and abundance (Balesic & Martin 1987; Brown et al. 1993;

Lynch et al. 2015). The wind roses displaying larval captures by sampling week (Figure

6.5) clarified the importance of week 2 onshore north and northeast winds in explaining the higher densities of ichthyoplankton in the study region. The time series of conventional wind direction roses showed that prevailing winds at Douglas Point were highly variable from week to week, but that week 2 was generally associated with more onshore winds than any other week sampled. Several investigators have previously discussed the importance of wind-induced surface currents in determining the transport and retention of ichthyoplankton in general (Hook et al. 2006; vanderVeer et al. 2009), and larval Lake Whitefish in particular (Wismer et al. 1986; Ryan & Crawford 2014;

Lynch et al. 2015). Specifically, Saylor and Miller (1976) reported that the Main Basin prominent counter-clockwise alongshore current is mainly driven by wind stress on surface waters. Drift of larval Lake Whitefish by the longshore current makes it almost impossible to identify spawning shoals and important nursery habitats (Balesic & Martin

1987). Based on these considerations, it would be extremely helpful to conduct more

178 detailed investigations on the specific effects of wind-induced surface currents on distribution and abundance of ichthyoplankton at Douglas Point.

Previous studies of the ecology of larval Lake Whitefish have indicated the importance of water temperature for successful embryonic development (i.e. Brooke

1975; Ryan & Crawford 2014; Eme et al. 2015), recruitment (i.e. Casselman 2002) and feeding (i.e. Roseman et al. 2005; Gobin et al. 2015) and indirectly through ice development (Christie 1963; Lawler 1965; Lynch et al. 2015). Water temperature was not identified as a significant environmental condition explaining the variation in the distribution of larval Lake Whitefish or other ichthyoplankton.

Serendipitous Samples of Aggregated Ichthyoplankton

On one hand, the personal observations and sampling of aggregated ichthyoplankton at physical structures in nearshore waters of the Douglas Point embayments during 2014 turned out to consist of species other than Lake Whitefish (i.e.

Longnose Sucker and White Sucker). However, it was the simple observation of such large larval aggregations that raised a few important questions for this investigation. First, were the larvae actually 'shoaling' = responding in like manner to common environmental cues, or were they 'schooling' = responding in a common, positive social interaction to each other (Faber 1970; Holmes et al. 2002)? Personal observation while attempting to sample the aggregations with a dip net suggested that the suckers were indeed schooling, since they avoided the gear while largely maintaining dynamic cohesion. This is consistent with previous investigations, where Campbell (1971) noted that White Suckers are often found swimming together in large groups during the day. Second, aggregation

179 of sucker larvae leads directly to the question of whether Lake Whitefish larvae might also be forming such aggregations, and whether larval schooling might be an important factor in finding and sampling these aggregations in the wild. Hart (1930) proposed that when Lake Whitefish larvae hatch, they form schools later in the larval period before migrating to much deeper waters. This suggestion is contradictory to Bajkov (1930), who observed that larval Lake Whitefish schooled early in the larval period, and to

Hoagman (1974), who suggested that Lake Whitefish were entirely non-schooling organisms. Zitzow and Millard (1988) hypothesized that larval Lake Whitefish only formed schools when startled. Faber (1970) found from his visual observations that larval Lake Whitefish did not school in the formal sense, but his peak abundance in surface water plankton tows was 61 larvae (Faber 1970). Thus, the shoaling or schooling of larval Lake Whitefish remains a key uncertainty that requires further investigation.

Finally, it is important to consider that all of the larval Lake Whitefish collected during this investigation were sampled exclusively in surface and/or nearshore waters.

The rather sparse literature on Lake Whitefish larval ecology is divided on the locomotory abilities and preferences of larvae with regard to exploiting deeper habitats.

Faber (1970) noted that larval Lake Whitefish tend to congregate in surface waters over variable water depths (1 to 10m deep), which was also seen in this study with four larval

Lake Whitefish sampled in the middle of Inverhuron Bay, two sampled in mid-Holmes

Bay and two sampled in the mouth of Holmes Bay. Although the majority of larval Lake

Whitefish were caught within the embayments at nearshore transects, it is important to note that these tows were sampled over the 2m depth contour. These observations stand in contrast to Reckahn (1970), who found larval Lake Whitefish only in water depths less

180 than 1m. While horizontal surface plankton tows have historically been the sampling standard for larval Lake Whitefish in the wild (e.g., Loftus 1978a; 1978b; Ryan &

Crawford 2014), these plankton tows are not necessarily the most efficient or effective method for finding and sampling targets that are (a) often aggregated, and/or (b) deeper than 1m, and/or (c) relatively mobile, and/or (d) socially responsive to perceived threats.

Taken together with the 2013 field study (Chapter 2) conclusions about the importance of

15m+ vertical diurnal migrations by zooplankton in the study region, the observations from the 2014 field study combine to focus attention on significantly enhancing our understanding of the three-dimensional spatial-temporal habits of these larval Lake

Whitefish and their zooplankton prey.

Hypotheses and Predictions

The goal of this investigation was to determine the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron. In order to achieve this goal, a 2014 field program was executed in a manner that allowed for the empirical evaluation of three hypotheses and associated predictions.

The Larval Pulse Hypothesis stated that spring hatching of Lake Whitefish, and corresponding peaks in spring primary/secondary production (zooplankton prey) typically associated with nearshore Great Lakes waters after ice-out, lead to a distinct pulse in larval abundance that reduces as ichthyoplankton deplete their prey resources (Faber

1970; Bidgood 1974, Peeters et al. 2007). This pulse in larvae reduces to lower levels as ichthyoplankton reduce availability of their prey resources (Faber 1970; Bidgood 1974).

181 If this hypothesis was true, then there would have been an increase in larval standard densities during the first weeks of sampling, followed by a decline during the later weeks

(e.g., weekly SD 1<2<3>4>5). Redundancy Analysis for the effect of space- time/environment did identify week as a significant factor explaining standard densities for larval Lake Whitefish (Tables 6.8, 6.9), and an examination of the time series showed a dramatic peak during week 2 of this study, followed by rapid decline to very low larval densities (Table 6.5, Figure 6.5). Based on these observations, the Larval Pulse

Hypothesis is moderately supported by this investigation - although data are still required to characterize the recruitment dynamics prior to ice-out. These observations are consistent with other larval Lake Whitefish sampling programs, where a distinct temporal pulse has been noted in early May (Faber 1970; Loftus 1978a; 1978b; Freeberg et al.

1990; Ryan & Crawford 2014).

The Longshore Transport Hypothesis stated that ichthyoplankton in general, and larval Lake Whitefish in particular, are consistently moved by the prevailing geostrophic longshore currents (Hoagman 1973; Freeberg et al. 1990; Claramunt et al. 2010a), which in the Douglas Point region of Lake Huron flow in a northerly direction 80% of the time

(Wismer et al. 1986). If this hypothesis was true, then there would have been a spatio- temporal sequence of high larval standard densities along a latitudinal gradient from south to north during the sampling period (i.e, Inverhuron Bay < Holmes Bay < Baie du

Doré). While Redundancy Analysis did identify week as a significant factor for larval standard densities, it did not identify bay as a significant factor (Table 6.8). No latitudinal gradient was observed for larval Lake Whitefish at Douglas Point in 2014 (Table 6.5), and thus there is no support for The Longshore Transport Hypothesis from this

182 investigation. However, it is important to note that the major effects of the artificial

BNGS A and B cooling water intakes/outflows on Lake Huron water currents (Balesic &

Martin 1987; Balesic & Kwik 1991) could easily have swamped any latitudinal gradient caused by natural alongshore transport of ichthyoplankton; especially in deeper 10-20m water around the intakes and in proximity to the major discharge flow extending more than a kilometer offshore (Holmes & Noakes 2002).

The Embayment Nursery Hypothesis stated that nearshore embayment morphology causes eddies in water currents that retain both ichthyoplankton and their zooplankton prey (Zhao et al. 2012; Howell et al. 2014). Nearshore embayments can retain both ichthyoplankton, including Lake Whitefish larvae, and their zooplankton prey

(Cowen 2002). If this hypothesis was true, then there would have been higher standard densities of larvae within the embayments, compared to samples from outside the embayments (e.g. Inverhuron Bay: Out

Redundancy Analysis did not identify either bay or offshore as significant factors explaining larval standard densities at Douglas Point in 2014 (Table 6.8, 6.9). It is possible that larval fish were present during sampling events but that they were able to avoid the sampling gear (Brander & Thompson 1989), or that larvae were not present due to mortality (e.g., starvation, predation; Rice et al. 1987; Ryan & Crawford 2014), or that larvae were not present due relocation to offshore deeper waters (Loftus 1978a; Cucin &

Faber 1985; McKenna & Johnson 2009). If larval Lake Whitefish transitioned from being passive planktonic organisms to being active prey-seeking individuals (McKenna

& Johnson 2009), larvae could have moved substantially further offshore over greater depths in the water column, with the possibility that they would pursue zooplankton

183 during diurnal vertical migration (Beeton 1960; Geller 1986; Lampert 1989) (see Chapter

2 of this thesis). However, it is worth noting that the three observations of high standard density for (schooling?) larval Lake Whitefish all occurred in nearshore waters (Table

6.5). Roseman et al. (2005) reported that nearshore waters were associated with higher abundance of planktonic food for larval fishes, and this is consistent with the majority of larval Lake Whitefish being caught in nearshore waters in this study (Table 6.5). If future investigations could significantly enhance the ability to sample larval aggregations - especially in deeper waters - it is possible that the Embayment Nursery Hypothesis could still be a viable contender to help explain the distribution and abundance of larval Lake

Whitefish at Douglas Point.

In conclusion, the results of this investigation have important implications for a variety of different interests and applications. Lake Whitefish are found in the source waters of Douglas Point and surrounding nearshore embayments during all stages of their life history (i.e., spawning, embryonic overwintering, larval development), and therefore continued monitoring of this species is essential, especially due to the operation of the

BNGS, North America’s largest nuclear power plant. As part of the CNSC requirements for the Whitefish Follow Up Monitoring Program, triggered in large part because of the

Constitutionally-protected Indigenous fishery in Lake Huron, Lake Whitefish have been sampled yearly; however, it is clear that assessment of larval Lake Whitefish in the source waters of Lake Huron has been largely ignored (Brown 2007; Fietsch 2011).

Understanding the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish is an essential step in mitigating the potential effects of the

184 BNGS on a critically important developmental period of this ‘Valued Ecosystem

Component’ (Holmes et al. 2002; Crawford et al. 2003; Bruce Power 2012).

185 6.6. References

Auer N. A. 1982. Identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage. Special Publication, 82–3. Great Lakes Fishery Commission, Ann Arbor, Michigan, USA.

Bajkov A. 1930. A study of the whitefish (Coregonus clupeaformis) in Manitoban Lakes. Contrib. Can. Biol. Fish. 5(15): 441-455.

Balesic H., Martin I.D. 1987. Bruce GS ‘B’ Preoperational Biological Study Summary, 1979 to 1983. Ontario Hydro Research Division Report N. 87-185-K. 43p.

Balesic H., Kwik J.K. 1991. Bruce GS ‘B’ Commissioning Aquatic Biological Studies Year IV, 1986, and 1983 to 1986 Summary. Ontario Hydro Research Division Report No. 91-63-K. 90p.

Beeton A. M. 1960. The vertical migration of Mysis relicta in Lakes Huron and Michigan. Journal of the Fisheries Research Board of Canada 17(4): 517-539.

Bidgood B. F. 1974. Reproductive potential of two lake whitefish (Coregonus clupeaformis) populations. Journal of the Fisheries Research Board of Canada 31: 1631- 1639.

Brander K., Thompson A. B. 1989. Diel differences in avoidance of three vertical profile sampling gears by herring larvae. Journal of Plankton Research 11(4): 775-784.

Brooke L.T. 1975. Effect of different constant incubaion temepratures on egg sirvival and embryonic development in lake whitefish (Coregonis clupeaformis). Transactions of the American Fisheries Society 3:555-559

Brown R. W., Taylor W. W. 1992. Effects of egg composition and prey density on the larval growth and survival of lake whitefish (Coregonus clupeaformis Mitchell). Journal of Fish Biology 40: 381-394.

Brown R.W., Taylor W. W., Assel R.A. 1993. Factors affecting the recruitment of lake whitefish in two areas of northern Lake Michigan. Journal of Great Lakes Research 19: 418-428.

Brown T. 2007. Whitefish Investigations 2006 Summary. B-REP-00531-00017. Bruce Power. 7pp.

Bruce Power. 2012. Operations Phase Impingement and Entrainment Monitoring Revised Plan, March 2012. Bruce Power, Tiverton, Ontario, Canada.

186 Campbell K. P. 1971. Influence of light and dark periods on spatial distribution and activity of the white sucker, Catostomus commersoni. Transactions of the American Fisheries Society 100(2): 353-355.

Casselman J.M. 2002. Effects of temperature, global extremes and climate change on year-class production of warmwater, coolwater, and coldwater fishes in the Great Lakes basin. American Fisheries Society Symposium 32:39-60.

Christie W.J. 1963. Effects of artificial propogation and the weather on recruitment in the Lake Ontario whitefish fishery. Fisheries Research Board of Canada 20: 597-646.

Clady M., Hutchinson B. 1975. Effect of high winds on eggs of yellow perch, Perca flavescens, in Oneida Lake, New York. Trans. Amer. Fish. Soc., 104:524-525.

Claramunt R. M., Muir A. M., Johnson J., Sutton T. M. 2010a. Spatio-temporal trends in the food habits of age-0 lake whitefish. Journal of Great Lakes Research 36: 66-72.

Claramunt R. M., Muir A.M., Sutton T.M., Peeters P.J., Ebener M.P., Fitzsimons J.D., Koops M.A. 2010b. Measures of larval lake whitefish length and abundance as early predictors of year-class strength in Lake Michigan. Journal of Great Lakes Research 36: 84-91.

CNSC. 2003. Record of proceedings, including reasons for decision in the matter of Bruce Power Inc. Environmental Assessment Screening Report for the return to service of Units 3 & 4 of the Bruce Nuclear Generating Station (NGS) ‘A.’ Canadian Nuclear Safety Commission, Ottawa, Ontario, Canada. 06 January 2003. 25+pp.

Cowen R. K. 2002. Larval dispersal and retention and consequences for population connectivity in Sale P. F. (eds) Coral reef fishes: dynamics and diversity in a complex ecosystem. Elsevier Science, USA.

Crawford S.S., Muir A., McCann K. 2001. Ecological basis for recommendation of 2001 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Report prepared for the Chippewas of Nawash First Nation, Wiarton, Ontario. 12 October 2001 (revised with references 11 July 2002; revised with response to OMNR comments 02 January 2003). 156+pp.

Crawford S., McCann K., Muir A. 2003. 2003 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Report prepared for the Chippewas of Nawash Unceded First Nation R.R. #5, Wiarton, Ontario N0H 2T0. 109pp.

Crawford S., Gillis D., Hanner R. 2014. University of Guelph Research Team Report, Year 3 Final Report: Saugeen Ojibway Nation-Bruce Power, UG-McMaster ‘Collaborative’ Whitefish Research Program. Saugeen Ojibway Nation-Bruce Power,

187 UG-McMaster ‘Collaborative’ Whitefish Research Program, University of Guelph, Guelph, Ontario, Canada. 49pp.

Cucin D., Faber D. J. 1985. Early life studies of lake whitefish (Coregonus clupeaformis, Mitchill), cisco (Coregonus artedii) and yellow perch (Perca flavescens) in Lake Opeongo, Ontario. Ontario Fisheries Technical Report Series, 16.

Ebener M. P., Brenden T. O., Wright G. M., Jones M. L., Faisal M. 2010. Spatial and temporal distributions of lake whitefish spawning stocks in Northern lakes Michigan and Huron, 2003-2008. Journal of Great Lakes Research 36: 38-51.

Ellis J. D. 1980. Preoperational Aquatic Biological Studies Bruce Nuclear Power Development 1973-1975. Report No. 80-291-K. 48 pp.

Eme J., Mueller C. A., Manzon R. G., Somers C. M., Boreham D. R., Wilson J. Y. 2015. Critical windows in embryonic development: shifting incubation temperatures alter heart rate and oxygen consumption of lake whitefish (Coregonus clupeaformis) embryos and hatchlings. Comparative Biochemistry and Physiology, Part A 179: 71-80.

ESG and Beak Internation Inc. 2000. Bruce Nuclear Power Development Ecological Effects Review Revision R00. Report G99242. 387p.

Faber, D. J. 1970. Ecological observations on newly hatched lake whitefish in South Bay, Lake Huron. Biology of Coregonid Fishes. C. C. a. C. S. W. Lindsey: 481-500.

Fietsch C-L. 2011. Whitefish Investigations 2010 Summary. Bruce Power Report. 9pp

Frank R. 1981. Pesticides and PCB in the Grand and Saugeen River Basins. Journal of Great Lakes Resaerch 7(4): 440-454.

Freeberg M. H., Taylor W. W., Brown R. W. 1990. Effect of egg and larval survival of year-class strength of lake whitefish in Grand Traverse Bay, Lake Michigan. Transactions of the American Fisheries Society 119: 92-100.

Gasiunaite Z. R., Cardoso A. C., Heiskanen A-S., Henriksen P., Kauppila P., Olenina I., Pilkaityte R., Purina I., Razinkovas A., Sagert S., Schubert H., Wasmund N. 2005. Seasonality of coastal phytoplankton in the Baltic Sea: Influence of salinity and eutrophication. Estuarine, Coastal and Shelf Science 65: 239-252.

Geller W. 1986. Diurnal vertical migration of zooplankton in a temperate great lake (L. Constance): A starvation avoidance mechanism? Arch. Hydrobiol. 1 (Suppl. 74): 1-60.

Gobin J., Lester N. P., Cottrill A., Fox M. G., Dunlop E. S. 2015. Trends in growth and recruitment of Lake Huron lake whitefish during a period of ecosystem change, 1985 to 2012. Journal of Great Lakes Research 41: 405-414

188 Golder Associates. 2007. Whitefish mark/recapture study 2006. Bruce Power. Report 05- 1112-082. 42pp.

Goodyear C. D., Edsall T. A., Ormsby Dempsey D. M., Moss G. D., Polanski P. E. 1982. Atlas of the spawning and nursery areas of Great Lakes Fishes Volume V: Lake Huron. NCE-IS-78-30. Report. 123pp.

Griffiths, J. S. 1974. Aquatic Biological Studies – Bruce Nuclear Power Development 1973. Ontario Hydro Research Division Report No. 74-449-H. 30p.

Hall T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser 41:95–98

Harrington M. M. 1895. Surface currents of the Great Lakes. Bull. B, U. S. Weath. Bur., Washington, D.C.

Hart J. L. 1930. The spawning and early life history of the whitefish, Coregonus clupeaformis (Mitchell), in the Bay of Quinte, Ontario. Contribut. Can. Biol. Fish NS 6(7):165-214

Hoagman W.J. 1973. The hatching, distribution, abundance, growth, and food of the larval lake whitefish (Coregonus clupeaformis Mitchill) of Central Green Bay, Lake Michigan. Institute of Freshwater Research, Drottningholm Report No. 53:1-20.

Hoagman W. J. 1974. Vital activity parameters as related to the early life history of larval and post-larval lake whitefish (Coregonus clupeaformis). Blaxter, J. H. S. (eds) in The early life history of fish. Springer-Verlag, New York, NY.

Holmes J. A., Noakes D. L. G. 2002. Whitefish Interactions with Nuclear Generating Station: Recommendations for Monitoring the Effects of the Bruce Nuclear Power Development on Whitefish in Lake Huron. Report. 173pp.

Holmes J. A., Noakes D. L. G., Crawford S. S., Wismer D. 2002. Whitefish Interactions with Nuclear Generating Stations. Aquatic Effects of Nuclear Generating Stations on Fishes. Report. 352 pp.

Home A. J., Goldman C. R. 1994. Limnology. New York: McGraw-Hill.

Hook T.O., McCormick M.J., Rutherford E.S., Mason D.M., Carter G.S. 2006. Short-term water mass movements in Lake Michigan: implications for larval fish transport. Journal of Great Lakes Research 32:728-737

Howell E. T., Barton D. R., Fietsch C-L., Kaltenecker G. 2014. Fine-scale nutrient enrichment and water quality on the rural shores of Southeast Lake Huron. Journal of Great Lakes Research 40: 126-140.

189 Hoyle J. A. 2003. Larval lake whitefish feeding and growth study, 2003. Ontario Ministry of Natural Resources, Lake Ontario Management Unit. Internal Report-LOA 03.03. 14pp.

Hoyle J.A., Johannsonn O. E., Bowen K. L. 2011. Larval lake whitefish abundance, diet and growth and their zooplankton prey abundance during a period of ecosystem change on the Bay of Quinte, Lake Ontario. Aquatic Ecosystem Health and Management 14(1): 66-74.

Hubert N., Hanner R., Holm E., Mandrak N.E., Taylor E., Burridge M., Watkinson D., Dumont P., Curry A., Bentzen P., Zhang J. April J., Bernatchez L. 2008. Identifying Canadian freshwater fishes through DNA barcodes. PlosOne 3(6):1-8.

Ivanova N.V., Zemlak T. S., Hanner R. H., Hebert P.D.N. 2007. Universal primer cocktails for fish DNA barcoding. Molecular Ecology Notes 7: 544-548.

Jude D. J., Tesar F. J., Tin H. T., 1998. Spring distribution and abundance of larval fishes in the St. Marys River, with a note on potential effects of freighter traffic on survival of eggs and larvae. Journal of Great Lakes Research 24(3): 569-581.

Kikko T., Ishizaki D., Ninomiya K., Kai Y., Fujioka Y. 2015. Diel patterns of larval drift of honmoroko Gnathopogon caerulescens in an inlet of Ibanaiko Lagoon, Lake Biwa, Japa. Journal of Fish Biology 86: 409-415.

Klumb R.A., Rudstam L. G., Mills E. L., Schneider C. P., Sawyko P. M. 2003. Importance of Lake Ontario embayments and nearshore habitats as nurseries for larval fishes with emphasis on alewife (Alosa pseudoharengus). Journal of Great Lakes Research 29(1): 181-198.

Lampert W. 1989. The adaptive significance of diel vertical migration of zooplankton. Functional Ecology 3: 21-27.

Lane J. A., Portt C. B., Minns C. K. 1996. Nursery habitat characteristics of Great Lakes fishes. Canadian Manuscript Report of Fisheries and Aquatic Sciences No. 2338:v+42p.

Lawler G.H. 1965. Fluctuations in the success of year-classes of lake whitefish populations with special reference to Lake Erie. Journal of the Fisheries Research Board of Canada 22:1197-1227.

Legendre P., Legendre L. 2012. Numerical Ecology 3rd Edition. Amsterdam, The Netherlands.

Leslie K. J. 1986. Nearshore contagion and sampling of freshwater larval fish. Journal of Plankton Research 8(6): 1137-1147.

Loftus D. H. 1978a. Early assessment of the 1977 year class of lake whitefish in Lake Huron. Lake Huron Fisheries Assessment Unit, OMNR, Owen Sound. Report 7pp.

190

Loftus D. H. 1978b. Larval fish sampling in Lake Huron, 1978. Lake Huron Fisheries Assessment Unit, OMNR, Owen Sound. Report 3-78. 32pp.

Lynch A. J., Taylor W. W., Bear T. D. Jr., Lofgren B. M. 2015. Climate change projections for lake whitefish (Coregonus clupeaformis) recruitment in the 1836 Treaty Waters of the Upper Great Lakes. Journal of Great Lakes Research 41: 415-422.

McKenna Jr. J. E., Johnson J. H. 2009. Spatial and temporal variation in distribution of larval lake whitefish in eastern Lake Ontario: Signs of recovery? Journal of Great Lakes Research 35: 94-100.Mills, K. H. 1985. Responses of lake whitefish (Coregonus clupeaformis) to fertilization of Lake 226, the Experimental Lakes Area. Can. J. Fish. Aquat. Sci. 42: 129-138.

McKinley R. S., Sim B. 1986. Assessment of ichthyoplankton entrainment at Bruce NGS ‘A’-1985. Ontario Hydro Research Division Report No. 86-190-K. 28pp.

Mills K. H. 1985. Responses of lake whitefish (Coregonus clupeaformis) to fertilization of Lake 226, the Experimental Lakes Area. Can . J. Fish. Aquat. Sci. 42:129-138.

Naaum A.M., Foottit R.G., Maw H.E.L., Hanner R. 2012. Differentiation Between Aphis pomi and Aphis Spiraecola Using Multiplex Real-Time PCR Based on DNA Barcode Sequences. Journal of Applied Entomology. 136(9):704-710.

Nabout J. C., Siqueira T., Bini L. M., Nogueira I de S. 2009. No evidence for environmental and spatial processes in structuring phytoplankton communities. Acta Oecologica 35: 720-726.

Newman D. G. 2014. Revisiting the Duty to Consult Aboriginal Peoples. Purich Publishing Ltd., Saskatoon, Saskatchewan, Canada. 184 pp.

O’Connell C. P., Raymond L. P. 1970. The effect of food density on survival and growth of early post yolk-sac larvae of the northern anchovy (Engraulis mordax Girard) in the laboratory. J. Exp. Mar. Biol. Ecol. 5:187-197.

OPG (Ontario Power Generation). 1999. Bruce 5-8 environmental effects report. Ont. Power Gen.-Nuclear, Tech. Sup. Environ. Aff. Subdiv., NK29-REP-07010.011-001R00.

Pan M,. McBeath A.J.A., Hay S.J. et al. 2008. Real-time PCR assay for detection and relative quantification of Liocarcinus depurator larvae from plankton samples. Marine Biology. 153:859-870.

Peeters F., Straile D., Lorke A., Ollinger D. 2007. Turbulent mixing and phytoplankton spring bloom development in a deep lake. Limnology and Oceanography 52(1): 286-298.

191 Peng S., Qin X., Shi H., Zhou R., Dai M., Ding D. 2012. Distribution and controlling factors of phytoplankton assemblages in a semi-enclosed bay during spring and summer. Marine Pollution Bulletin 64: 941-948.

Pothoven S. A., Hook T. O., Roswell C. R. 2014. Feeding ecology of age-0 lake whitefish in Saginaw Bay, Lake Huron. Journal of Great Lakes Research Supplement 40: 148-155.

Pritt J.J., Roseman E. F., Ross J. E. 2015. Using larval fish community structure to guide long-term monitoring of fish spawning activity. North American Journal of Fisheries Managament 35: 241-252.

Ratnasingham S., Hebert P. D. N. 2007. BOLD: The Barcode of Life Data System (www.barcodinglife.org). Molecular Ecology Notes 7: 355-364.

Reckahn, J. A. 1970. Ecology of young lake whitefish (Coregonus clupeaformis) in South Bay, Manitoulin Island, Lake Huron. Biology of Coregonid Fishes. C. C. Lindsey and C. S. Woods.

Rice J. A., Crowder L. B., Binkowski F. P. 1987. Evaluating potential sources of mortality for larval bloater (Coregonus hoyi): starvation and vulnerability to predation. Canadian Journal of Fisheries and Aquatic Science 44: 467-472.

Roseman E.F., Taylor W.W., Hayes D.B., Tyson J.T., Haas R.C. 2005. Spatial patterns emphasize the importance of coastal zones as nursery areas for larval walleye in Western Lake Erie. Journal of Great Lakes Research 31(1): 28-44

Roseman E. F., Kennedy G., Manny B. A., Boase J., McFee J. 2012. Life history characteristics of a recovering Lake Whitefish Coregonus clupeaformis stock in the Detroit River, North America. Advanc. Limnol. 63: 477-501.

Ryan K. M., Crawford S. S. 2014. Distribution and abundance of larval lake whitefish (Coregonus clupeaformis) in Stokes Bay, Lake Huron. Journal of Great Lakes Research 40(3): 755-762.

Saylor J. H., Miller G. S. 1976. Winter currents in Lake Huron. NOAA Tehcnical Memorandum ERL GLERL-15. Environmental Protection Agency Report No. EPA- 905/4-75-004. 118pp.

Sloss P. W., Saylor J. H. 1975. Measurements of current flow during summer in Lake Huron. NOAA Technical Report ERL 353-GLERL. GLERL Contribution No. 56. 45pp.

Swearer S. E., Shima J. S. 2010. Regional variation in larval retention and dispersal drives recruitment patterns in a temperate reef fish. Marine Ecology Progress Series 417: 229-236.

192 Taylor W. W., Freeberg M. H. 1984. Effect of food abundance on larval lake whitefish, Coregonus clupeaformis Mitchell, growth and survival. Journal of Fish Biology 25: 733- 741.

Taylor W. W., Smale M. A., Freeberg M. H. 1987. Biotic and abiotic determinants of Lake Whitefish (Coregonus clupeaformis) recruitment in Northeastern Lake Michigan. Canadian Journal of Fisheries and Aquatic Science 44(Suppl. 2): 313-323.

Ter Braak, C. J. F., Smilauer P. 2012. Canoco reference manual and user’s guide: software for ordination (version 5.0). Microcomputer Power, Ithaca, NY. USA. vanderVeek H. W., Bolle L. J., Geffen A. J., Witte J. I. J. 2009. Variability in transport of fish eggs and larvae. IV. Interannual variability in larval stage duration of immigrating plaice in the Dutch Wadden Sea. Marine Ecology Progress Series 390: 213-223.

Ward R.D., Hanner R., Hebert P.D.N. 2009. The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology 74: 329-356.

Wismer D. A., Hoffer A. D., Chubbuck D. A. 1986. Bruce B Generating Station Pre- Operational Environmental Summary 1975-1983. Environmental Studies and Assessments Department Report No. 86246. 95pp.

Xuemei L., Yuhe Y., Tanglin Z., Weisong F., Hongyi A., Qingyun Y. 2012. Seasonal variation of plankton communities influenced by environmental factors in an artificial lake. Chinese Journal of Oceanology and Limnology 30(3): 397-403.

Zhao J., Rao Y. R., Sheng J. 2012. Numerical study of dispersion and hydrodynamic connectivity of near-surface waters in Lake Huron. Water Quality Research Journal of Canada 47: 3-4.

Zitzow R. E., Millard J. L. 1988. Survival and growth of lake whitefish (Coregonus clupeaformis) larvae fed only formulated dry diets. Aquaculture 69: 105-113.

193

Figure 6.1.A-B. Location of transects (lines), pump stations (stars) of (A) Inverhuron Bay and Holmes Bay and (B) Baie du Doré at Douglas Point, Lake Huron in 2014. Classification of transects correspond to Table 6.1.

194

Figure 6.2. Biplot of the Redundancy Analysis after a forward selection of the relationship between space-time variables on environmental conditions sampled at Douglas Point, Lake Huron in 2014. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3.

195

Figure 6.3A-C. Time series of temperature (°C) from data loggers at 2m depth on nearshore transects in (A) Inverhuron Bays and (B,C) Baie du Doré, Douglas Point, Lake Huron, for the period May-September 2014.

196

Figure 6.4. Uniplot of the Redundancy Analysis after a forward selection of the relationship between space-time/environment on ichthyoplankton standard densities sampled at Douglas Point, Lake Huron in 2014. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3.

197

Figure 6.5. Standard densities (SD) of ichthyoplankton (# larvae 1000m-3) sampled in 2014 at Douglas Point, Lake Huron, associated with prevailing=modal) wind direction on the day of sampling. (A) Lake Whitefish (Coregonus clupeaformis), (B) all species. Open circles represent observations with SD<10, while bold numbers represent week number.

198

Figure 6.6. Weekly wind roses for direction data collected at Gunn Point, Inverhuron, Lake Huron for the six-week period from 5 May - 21 June 2014. Bold pies in sectors represent percentage of wind station observations associated with wind moving toward the direction indicated for the sector. Note that all weeks were 7 days except for week 1 which was 11 days.

199 Table 6.1. Classification of transects by region, bay and offshore designations. Transects can be classified as in embayment or out of embayment (Bay); near or offshore (Offshore). Associated variable types correspond to space-time variables used in Redundancy Analyses. Transect codes correspond to Figure 6.1A-B.

Region Bay Offshore Transect Inverhuron In Near IH3 IH2 IH4 IH5 Far 1H6 Out Near IH1 Far 1H7 1H8 Holmes In Near H1 H3 Far H2 Out Near H5 Far H4 Baie du Doré In Near B4 B5 BP1 BP2 Out Near B3 Far B1 B2

200 Table 6.2. Dates of plankton samples for each of the three nearshore embayments at Douglas Point, Lake Huron during the 2014 field season.

Bay Week Inverhuron Bay Holmes Bay Baie du Doré 1 5-7, 11 May 8,11 May 8, 12 May 2 19, 22 May 21 May 20, 21 May 3 26, 29 May 26 May 27, 30 May 4 5-8 June 7 June 6 June 5 9-10, 13 June 13-14 June 10 June 6 18-21 June 19 June 20-21 June

201 Table 6.3. Space-time variables, environmental conditions and response variables sampled at Douglas Point, Lake Huron in 2014 with associated codes referenced in Redundancy Analyses tables and figures.

Variable Type Variable Names Units Code Spatial Region Inverhuron Bay IH Holmes Bay HO Baie du Doré BD Bay In embayment IN Out of embayment OUT Offshore Nearshore NEAR Offshore FAR Temporal Week Week 1 Wee1 Week 2 Wee2 Week 3 Wee3 Week 4 Wee4 Week 5 Wee5 Week 6 Wee6 Environmental Temperature (log(x+1)) °C L-Temp Specific conductivity µs cm-1 SCond Dissolved oxygen µg L-1 DO pH pH Wind speed kph WindS8 Wind Direction East E South West SW North West NW South S South East SE North East NE North N West W Response Standard density of other # larvae larval fish collected 1000m -3 SD-OTHER Standard density of larval # larvae Lake Whitefish collected 1000m-3 SD-WHF

202 Table 6.4. Consistency of Lake Whitefish (Coregonus clupeaformis) specific Real-time PCR (qPCR) compared to DNA barcode species identification of all ichthyoplankton caught at Douglas Point, Lake Huron in 2014. Real-time PCR (qPCR) threshold cycle (Ct) value indicates a relative measure of target species in a sample, where: Ct<30 indicates occurrence of target Lake Whitefish, Ct≥30 indicates possible occurrence of target; Ct=NA ‘not applicable’ indicates absence of target. Vis ID = consistency of visual identification using a morphological key to species identified by DNA barcoding.

DNA Barcoding Identification Family Unknown Salmonidae Catastomidae Cyprinidae Genus Unknown Coregonus Prosopium Catostomus Cyprinus Species Unknown clupeaformis artedi cylindraceum catostomus commersoni carpio Subtotal NA 22 34 1 57 qPCR >30 1 1 1 7 4 14 Ct <30 43 43 value Subtotal 1 43 1 1 29 38 1 114

Visual ID Consistency 0% 74%a 0%b 100% 0%c 100% 0%d a 32 identified as Lake Whitefish, 8 identified as Lake Herring and 3 identified as White Sucker b 1 identified as Lake Whitefish c 8 identified as Lake Whitefish, 21 identified as White Sucker d 1 identified as Fathead Minnow

203 Table 6.5. Distribution and abundance of larval Lake Whitefish (Coregonus clupeaformis) observed at Douglas Point, Lake Huron in 2014. Larval abundance and associated standard densities (# larvae 1000m-3 in parentheses). Transect codes (IH,H,B#) are associated with Table 6.1.

Region Bay Offshore Week Subtotal 1 2 3 4 5 6 Inverhuron In Near IH3: 11(25.7) IH3: 1(3.3) 28 IH2: 4(3.3) IH2: 1(2.9) IH4: 10(52.5) IH5: 1(7.3) Far IH6:4(9.7) 4 Out Near IH1:1(3.1) IH1:1(4.2) 2 Far Holmes In Near Far H2: 2(8.2) 2 Out Near H5:4(21.1) 4 Far H4:2(6.5) 2 Baie du In Near Doré Far Out Near B3:1(5.6) 1 Far Subtotal 7 32 2 1 1 43

204 Table 6.6. Axis summary statistics from the Redundancy Analysis of space-time variables on environmental conditions observed at Douglas Point, Lake Huron in 2014. Correlations of environmental conditions and ordination axes. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3. NS=not selected by analysis.

Statistics Axis 1 Axis 2 Space-time Variables Week Week 2 -0.1315 -0.3937 Week 3 -0.1859 -0.3184 Week 4 0.5096 0.176 Week 5 -0.1112 -0.528 Week 6 0.4218 0.2771 Wee1 -0.5869 0.7122 Bay BD 0.5548 0.0346 HO -0.1011 -0.1433 IH -0.4496 0.09 Bment IN NS NS OUT NS NS Offshore FAR NS NS NEAR NS NS

Environmental Conditions Wind Speed WindS8 -0.3684 0.0418 Wind Direction W -0.4825 0.6414 SE -0.0421 -0.5392 N 0.0844 -0.1964 NW 0.0769 0.0655 SW 0.2366 0.0956 S 0.248 0.1575 NE 0.009 0.0526 Ltemp 0.8484 0.0293 Scond 0.5711 0.0075 DO -0.6608 -0.4361 pH 0.325 -0.2022

Axis Summary Eigenvalues 0.1757 0.0846 Pseudo-canonical correlation 0.8806 0.8543 Explained variation (cumulative) 17.57 25.87 Explained fitted variation 43.32 67.82 (cumulative) F 20.50 11.30 p 0.001 0.001

205 Table 6.7. Variance partitioning analysis of the significant space-time factors on environmental conditions at Douglas Point, Lake Huron in 2014, as determined by Redundancy Analysis forward selection. Axis loadings and space-time variable/environmental condition abbreviations correspond to Table 6.3.

Factor Factor % F p (whole) Explains Week Week1 10.8 12.6 0.001 Week6 8.1 10.3 Week4 8.5 11.9 Week5 1.6 2.3 Week2 0.5 0.7 Week3 0.5 0.7 Bay BD 7.7 12.1 0.001 IH 1.0 1.6 HO 1.0 1.6

206 Table 6.8. Axis summary statistics from the Redundancy Analysis of space- time/environmental conditions on ichthyoplankton standard densities sampled at Douglas Point, Lake Huron in 2014. Space-time variable and environmental condition abbreviations correspond to those in Table 6.3. NS=not selected by analysis.

Statistics Axis 1 Axis 2 Space-Time Variables Week Wee2 0.6542 -0.0724 Wee3 -0.0376 0.2149 Wee4 -0.0923 0.2302 Wee5 -0.2894 -0.1929 Wee6 -0.1306 0.2272 Wee1 -0.1214 -0.3515 Bay BD NS NS HO NS NS IH NS NS Bment IN NS NS OUT NS NS Offshore FAR NS NS NEAR NS NS Environmental Conditions Wind Speed WindS8 NS NS Wind Direction W -0.2224 -0.0521 SE -0.2064 0.0103 N 0.074 0.5152 NW -0.0422 0.056 SW -0.2752 -0.1267 S -0.132 -0.0402 NE 0.6741 -0.3202 Ltemp NS NS Scond NS NS DO NS NS pH NS NS Response Variables SD-OTHER 0.4360 0.2301 SD-WHF 0.4502 -0.2228 Axis Summary Eigenvalues 0.2334 0.0691 Pseudo-canonical correlation 0.6622 0.3845 Explained variation (cumulative) 23.34 23.34 Explained fitted variation (cumulative) 77.15 100.0 F 25.90 8.4 p 0.005 0.845

207 Table 6.9. Variance partitioning analysis of the significant space-time/environment on ichthyoplankton standard densities at Douglas Point, Lake Huron in 2014, as determined by Redundancy Analysis forward selection. Axis loadings and space-time variable/environmental condition abbreviations correspond to Table 6.3.

Factor Factor % Explains F P (whole) Wind NE 9.5 10.9 0.001 Direction N 2.1 2.5 SW 0.2 0.2 W <0.1 0.1 S <0.1 <0.1 SE <0.1 <0.1 NW <0.1 <0.1 Week Week 2 6.7 8.1 0.001 Week 6 3.3 4.2 Week 1 1.5 1.8 Week 4 0.9 1.2 Week 5 0.4 0.5 Week 6 0.4 0.5

208 6.7. Appendices

Appendix 6.1. Non-Lake Whitefish ichthyoplankton distribution and abundance.

Table A6.1. Distribution and abundance of non-Lake Whitefish ichythoplankton caught at Douglas Point, Lake Huron in 2014. excluding larval Lake Whitefish. Abundance and standard densities (# larvae 1000m-3 in parentheses) provided. Subtotals provided per week and per region of each of the three bays; Inverhuron Bay, Holmes Bay and Baie du Dore.

Region Bay Offshore Week Subtotal 1 2 3 4 5 6 Inverhuron In Near IH3: IH3: 1(2.3) IH4:1(6.9) IH2: 3(6.4) IH4: 5(49.51) 22 1(3.6) IH2: 2(1.6) IH4:2( 10.9) IH4: 1(5.2) IH5:6(43.8) Far IH6:2(4.9) IH6: 1(3.6) IH6:1(1.98) 4 Out Near IH1:7(23.2) IH1:1(4.8) 8 Far IH7:1(1.3) IH8:1(4.7) 2 Holmes In Near H3:1(4.7) 1 Far Out Near H5:1(6.3) H5:1(5.8) 2 Far H4: 1(3.6) 1 Baie du In Near B5:2(13.3) B5:1(7.6) B5: 1(16.8) 12 Dore B4: 2(5.7) B4:5( 14.4) BP1:1( 20.4) Far Out Near B3:1(4.2) B3: 14(64.0) 15 Far B1: B1: 2(13.6) 4 2(4.3) Subtotal 4 28 20 13 0 6 71

209 Chapter 7. Epilogue

The ecology of larval Lake Whitefish (Coregonus clupeaformis) remains a major uncertainty in the study of Laurentian Great Lakes ecosystems, especially in the complex and dynamic environment of Douglas Point, Lake Huron. With continued operation of the Bruce Nuclear Generating Station (BNGS) - North America’s largest nuclear power plant, and the Deep Geological Repository for Nuclear Waste proposed for construction over the next decade, it is clear that federal Environmental Assessments (EA) on this species as a ‘Valued Ecosystem Component’ will require a much deeper understanding than is currently possessed. The goal of this thesis was to investigate potential relationships of environmental conditions on the ecology of larval Lake Whitefish at

Douglas Point, Lake Huron by addressing both outstanding ecological and methodological key uncertainties over five substantive chapters.

The goal of Chapter 2 was to determine which environmental conditions have a major effect on the distribution and abundance of zooplankton at Douglas Point, Lake

Huron. The major conclusion of this investigation was that two conditions, week

(temporal) and water sample depth (spatial), significantly explained the majority of variation observed in zooplankton density. Dissolved oxygen was identified as a significant environmental condition in the analysis, while environmental conditions such as temperature and pH were not major factors. While a temporal dimension was associated with secondary production biomass, the distribution and abundance of zooplankton appears to be much more complex than explained by the Seasonal Nutrient

Depletion Hypothesis. The strong positive association of zooplankton density and the

15m sample depth agreed with the Diurnal Migration Hypothesis, which is based on the

210 well-accepted phenomenon of zooplankton vertical migrations to deeper waters during daylight hours (Beeton 1960; Lampert 1989; Pannard et al. 2015). Given the extraordinary importance of zooplankton prey for the survival and development of larval

Lake Whitefish (Taylor & Freeberg 1984; Claramunt et al. 2010; Pothoven et al. 2014), it seems clear that future investigations will need to shift to the prey-tracking behaviour of the larvae that are potentially responding to vertically migrating prey (Dabrowski 1989;

Harris 1992). Despite the fact that many investigators have previously focused on nearshore embayments as prime nursery habitat (Freeberg et al. 1990; Klumb et al. 2003;

Claramunt et al. 2010), the current investigation suggests that attention should also be placed on larval Lake Whitefish in subsurface, offshore waters. The fact that zooplankton were observed in highest densities at 15m water depth during the day is important in two regards. First, nocturnal sampling may be an essential requirement to test the prediction that surface larval densities would be highest at night when the zooplankton prey return in their daily migration to the upper water column (Stocco & Joyeux 2015). Second, the results of this investigation suggest that both zooplankton and any associated prey- tracking larval fishes could be at substantial risk to entrainment from the cooling water intakes for BNGS A and B facilities - both of which are located less than 1km offshore at a depth of approximately 15m (Holmes & Noakes 2002). Given the water velocities generated by these industrial intakes, it is not clear to what extent the entrained organisms would still be recognizable for assessment after being removed from the source waters

(Cada & Hergenrader 1978; Ager et al. 2006; Paine et al. 2007). Clearly, the results of this investigation have several ramifications for the future of plankton monitoring

211 programs for the ongoing Douglas Point environmental assessments - specifically for ichthyoplankton, including larval Lake Whitefish.

The goal of Chapter 3 was to evaluate and explore the consistency between DNA barcoding and visual identification methods using a case study of larval fish caught in plankton tows at Stokes Bay, Lake Huron. This study is one of a handful of recent studies to specifically compare the accuracy and precision of visual and DNA barcoding identification methods for ichthyoplankton (i.e., Ko et al. 2013; Becket et al. 2015) and is the first such study to undertake a comparison with fishes from the Laurentian Great

Lakes. Accurate and precise identification of ichthyoplankton is essential for fish ecology, fisheries management, conservation and environmental assessments (Moser &

Smith 1993; Becker et al. 2015); however, visual identification alone can be highly unreliable (Pfrender et al. 2010; Ko et al. 2013). It is often difficult for analysts to visually identify embryos and larvae because of their size, lack of distinct visual characteristics, and broad overlap between morphological conditions of different species.

For example, consider the great phenotypic similarity between larval Lake Herring

Coregonus artedi and larval Lake Whitefish C. clupeaformis (see Auer 1982; Cucin &

Faber 1985). In this investigation, visual identification was actually found to be highly reliable when identifying to the level of family; however, error rates increased with progression to genus and species. This pattern of increasing error rates with key decision nodes supports both the Node Complexity Hypothesis and the Taxonomic Relatedness

Hypothesis. DNA barcoding provided an invaluable cross-reference for visual identification. However, using such genetic techniques incurs a different cost-benefit structure to conventional identification tools - especially in regards to technology,

212 reagents, training and access to reference databases (Moritz & Cicero 2004; Hebert &

Gregory 2005). For these reasons, we recommend a mixed-method approach to larval species identification for several reasons: issues of accuracy and precision can be estimated differently; cost-benefits can become more favorable; and because much can be learned when using a comparative approach to confront morphological and genetic data types, from the same specimens. With specific reference to environmental assessments such as the BNGS, it is clear that a combination of visual and DNA barcoding identification of ichthyoplankton has the potential to revolutionize the standard methods required for compliance with federal monitoring regulations.

The goal of Chapter 4 was to investigate basin-level haplotype variation and potential cryptic diversity of Lake Whitefish in Lake Huron, by expanding the barcode library across known spawning sites and times. In order to appropriately apply DNA barcoding and other genetic identification techniques as a tool for species identification, there is often a need to improve the sampling coverage of the target species, and to increase our knowledge of the genetic variation present both within and between species

(Meier et al. 2008; Knebelsberger et al. 2014). Previously it was thought that a sample size of five individuals was enough for accurate species identification using DNA barcoding (Hanner et al. 2005), but recently it has been suggested that greater spatial coverage and increased sample size are necessary to account for potential variation within a species - including the possibility of cryptic lineages (Young et al. 2013). Using a newly-compiled, comprehensive inventory of tissue from spawning-phase Lake

Whitefish sampled at numerous locations throughout all three basins of Lake Huron, it was possible to use a standard protocol of DNA barcoding to search for evidence of

213 cryptic lineages in this species/range. No such evidence was found for Lake Whitefish from Lake Huron, supporting the Panmixis Hypothesis, but not Basin Segregation

Hypothesis or the Sympatric Spawning Hypothesis. The occurrence of unique haplotypes among late-stage spawners could suggest a temporal, rather than a spatial trend in cryptic lineages; however, this possibility seems improbable in this case, given the co-occurrence of a dominant haplotype. Finally, characterizing potential haplotypes was a necessary first step in the successful design and implementation of a species-specific Lake

Whitefish qPCR assay, as described in the following Chapter.

Previously, there have been numerous research/management investigations of

Lake Whitefish population structure in the Laurentian Great Lakes (Ebener et al. 2010), especially for populations in Lakes Michigan and Ontario (Imhof et al. 1980; Ihssen et al.

1985, VanDeHey et al. 2010; Stott et al. 2010). These investigations have been highly variable in terms of spatial and temporal scope, ranging from local to basin-wide samples and from single years to longer time series (Casselman et al. 1981; Stott et al. 2010;

2013). With specific reference to Lake Whitefish population(s) in Lake Huron, population discrimination is highly uncertain for a variety of reasons, including: incomplete spatial sampling over non-randomly distributed spawning habitat, conflicting results from different population discrimination methods, evidence of basin-wide+ spawning migrations, and the very real possibilities that (a) significant numbers of mature individuals may not spawn in a given year, and (b) different populations may use common spawning grounds at different times during the spawning season (Loftus 1980,

Goodyear et al. 1982, Crawford et al. 2001). With specific reference to this thesis, it is important to note that rocky shoals in the greater Douglas Point region - like those near

214 Kettle/Stony Point and Point Clarks - are rather rare in southern Main Basin, despite the fact that the vast majority of the Lake Whitefish commercial fishery harvests are taken from that end of the lake (Ebener et al. 2008). Population discrimination has emerged as the single greatest ecological uncertainty associated with fisheries and EA management decision-making regarding Lake Whitefish in the Main Basin of Lake Huron, especially regarding the scaling of ecological effects associated with the BNGS and proposed Deep

Geologic Repository at Douglas Point (Bruce Power 2012; Swanson et al. 2015). While this uncertainty is clearly important for the ultimate interpretation of results from the current investigations, the complexity of the issue is well beyond the scope of this thesis.

The goal of Chapter 5 was to determine if a real-time PCR genetic primer/probe assay could be designed to identify Lake Whitefish using the DNA barcode region of the target species. DNA barcoding and other genetic techniques have previously been effective in identifying Lake Whitefish (e.g., allozymes-Casselman et al. 1981; microsatellites-Lu et al. 2001; restriction fragment length polymorphisms-Bernatchez &

Dodson 1991). These genetic methods provide some benefits over visual identification, but can also be costly and time-consuming for high-volume implementation.

Specifically, this chapter sought to develop a novel genetic method for reliably identifying larval Lake Whitefish in large, mixed samples of other ichthyoplankton - such as might be collected using plankton tows in the wild. qPCR was very effective at detecting larval Lake Whitefish in this study, with successful identification in samples with as little as 0.01% target DNA concentration. qPCR offers some substantial benefits over DNA barcoding because identification of a target species can be done in real time.

Certain qPCR platforms (e.g., Smart Cycle II) have also been designed for portability,

215 and so can be used in the lab and field. With specific reference to identification of larval fishes as part of environmental assessments at facilities such as BNGS, it should be noted that large-scale entrainment monitoring often leads to collection of 'mashes' - numerous bits and pieces of ichthyoplankton, some of which can be highly physically and/or biologically degraded (Cada & Hergenrader 1978). qPCR can identify fragments of a target species in a mash where visual identification would otherwise fail (Quinteiro et al.

1998; Mackie et al. 1999). Finally, it should be noted that the qPCR assay developed in this study was specifically intended for use as an empirical 'proof-of-concept' in the 2014 larval Lake Whitefish field investigation at Douglas Point, as described below.

Finally, the goal of Chapter 6 was to determine the effects of environmental conditions on the distribution and abundance of larval Lake Whitefish in nearshore embayments at Douglas Point, Lake Huron. Based on the results of this study, it was concluded that week and wind direction were the only significant factors explaining the variation seen in the distribution and abundance of ichthyoplankton in the three nearshore embayments: Inverhuron Bay, Holmes Bay and Baie du Doré. There was a very strong role played by peak abundance during week 2 of the six-week sampling period, in combination with onshore, northeasterly winds. These data provided support for the

Larval Pulse Hypothesis but not for the Longshore Transport Hypothesis or the

Nearshore Protection Hypothesis. The observations of three relatively high standard densities observed for larval Lake Whitefish all occurred in nearshore waters, raising the possibility that these larval aggregations reflected larval schooling behaviour - a phenomenon which has been previously hypothesized and reported for this species (Hart

1930; Bajkov 1930). If larval Lake Whitefish in the Douglas Point region (both nearshore

216 embayments and offshore waters) do in fact exhibit schooling behaviour, this factor would require a complete re-evaluation of previous ichthyoplankton sampling efforts for the BNGS environmental assessments. Locating and sampling schools of larvae can be exceedingly challenging in the field, and could have resulted in serious underestimation of abundance for larval Lake Whitefish by previous studies in the waters surrounding

Douglas Point.

With the recent relicensing of the Bruce Nuclear Generating Station and the proposal to place a Deep Geological Repository at the site (Bruce Power 2012; Swanson et al. 2015), the effects of environmental conditions on the distribution and abundance of plankton, including larval Lake Whitefish, at Douglas Point will continue to be highly relevant for many years to come. Hopefully, the contributions of this thesis will help to inform negotiations among Federal Regulators (Canadian Nuclear Safety Commission,

Fisheries and Oceans Canada, Environment Canada), Industry (Bruce Power, Ontario

Power Generation) and the affected First Nations (Chippewas of Nawash Unceded First

Nation, Saugeen First Nation) to help them make wise and effective decisions.

217 7.2. References

Ager D. D., Cord I., Patrick P. H. 2006. Entrainment sampling at Darlington Nuclear Generating Station – 2004. Report Number NK38-REP-07264-10001-R00.

Auer N.A. (ed.) 1982. Identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage. Great Lakes Fishery Commission, Ann Arbor, MI 48105. Special Pub. 82-3:744 pp.

Bajkov A. 1930. A study of the whitefish (Coregonus clupeaformis) in Manitoban Lakes. Contrib. Can. Biol. Fish. 5(15): 441-455.

Becker R. A., Sales N. G., Santos G. M., Santos G. B., Carvalho D. C. 2015. DNA barcoding and morphological identification of neotropical ichthyoplankton from the Upper Parana and Sao Francisco. Journal of Fish Biology. doi:10.1111/jfb.12707

Beeton A. M. 1960. The vertical migration of Mysis relicta in Lakes Huron and Michigan. Journal of the Fisheries Research Board of Canada 17(4): 517-539.

Bernatchez L., Dodson J.J. 1991. Phylogeographic structure in mitochondrial DNA of the Lake Whitefish (Coregonus clupeaformis) in North America and its relationships to Pleistocene glaciations. Evolution 45(4): 1016-1035.

Bruce Power. 2012. Operations Phase Impingement and Entrainment Monitoring Revised Plan, March 2012. Bruce Power, Tiverton, Ontario, Canada.

Cada G. F., Hergenrader G. L. 1978. An assessment of sampling mortality of larval fishes. Transactions of the American Fisheries Society 107 (2).

Casselman, J.M., Collins, J.J., Crossman, E.J. et al. 1981. Lake whitefish (Coregonus clupeaformis) stocks of the Ontario waters of Lake Huron. Canadian Journal of Fisheries and Aquatic Science, 38, 1772-1789.

Claramunt R. M., Muir A. M., Sutton T. M., Peeters P. J., Ebener M. P., Fitzsimons J. D., Koops M. A., 2010. Measures of larval lake whitefish length and abundance as early predictors of year-class strength in Lake Michigan. Journal of Great Lakes Research 36: 84-91.

Cucin, D., Faber, D.J. 1985. Early life studies of lake whitefish (Coregonus clupeaformis), cisco (Coregonus artedii) and yellow perch (Perca jlavescens) in Lake Opeongo, Ontario. Ont. Fish. Tech. Rep. Ser. No. 16.

Crawford S.S., Muir A., McCann K. 2001. Ecological basis for recommendation of 2001 Saugeen Ojibway commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. Report prepared for the Chippewas of Nawash First Nation, Wiarton, Ontario. 12 October 2001 (revised with references 11 July 2002; revised with response to OMNR comments 02 January 2003). 156+pp.

218 Dabrowski K. 1989. Formulation of a bioenergetic model for Coregonie early life history. 12th Annual Larval Fish Conference. Transactions of the American Fisheries Society 118: 138-150.

Ebener M. P., Kinnunen R. E., Schneeberger P. J., Mohr L. C., Hoyle J. A., Peeters P. 2008. Management of commercial fisheries for lake whitefish in the Laurentian Great Lakes of North America in: Schechter M. G., Leonard N. J., Taylor W. W. (eds.) International Governance of Fisheries Ecosystems: Learning from the Past, Finding Solutions for the Future. American Fisheries Society Press, Bethesda, Maryland, pp. 99–143.

Ebener M. P., Brenden T. O., Wright G. M., Jones M. L., Faisal M. 2010. Spatial and temporal distributions of lake whitefish spawning stocks in Northern lakes Michigan and Huron, 2003-2008. Journal of Great Lakes Research 36: 38-51.

Freeberg, M. H., Taylor, W. W., Brown, R. W. 1990. Effect of egg and larval survival on year-class strength of Lake Whitefish in Grand Traverse, Lake Michigan. Transactions of the American Fisheries Society 119: 92-100.

Goodyear C. D., Edsall T. A., Ormsby Dempsey D. M., Moss G. D., Polanski P. E. 1982. Atlas of the spawning and nursery areas of Great Lakes Fishes Volume V: Lake Huron. NCE-IS-78-30. Report. 123pp.

Hanner R., Schindel D., Ward B., Hebert P. 2005. Fish barcode of life (FISH-BOL). 1– 19.

Harris K. C. 1992. Techniques used for the fully-intensive culture of lake whitefish (Coregonus clupeaformis) larvae and yearlings in Ontario, Canada. Pol. Arch. Hydrobiol. 39(3/4*): 93-100.

Hart J. L. 1930. The spawning and early life history of the whitefish, Coregonus clupeaformis (Mitchell), in the Bay of Quinte, Ontario. Contribut. Can. Biol. Fish NS 6(7):165-214.

Hebert P. D. N., Gregory T. R. 2005. The promise of DNA barcoding for taxonomy. Systematic Biology 54(5): 852-859.

Hoagman W. J. 1974. Vital activity parameters as related to the early life history of larval and post-larval lake whitefish (Coregonus clupeaformis) in Blaxter J. H. S. (ed.). The early life history of fish. Springer-Verlag, New York, NY.

Holmes J. A., Noakes D. L. G. 2002. Whitefish Interactions with Nuclear Generating Stations. Recommendations for monitoring the effects of the Bruce Nuclear Power Development on Whitefish in Lake Huron. Report. 173 pp.

Ihssen P. E., Evans D. O., Christie W. J., Reckahn J. A., DesJardine R. L. 1981. Life history, morphology, and electrophoretic characteristics of five allopatric stocks of lake

219 whitefish (Coregonus clupeaformis) in the Great Lakes Region. Canadian Journal of Fisheries and Aquatic Science 38: 1790-1807.

Imhof M., Leary R., Booke H. E. 1980. Population or stock structure of lake whitefish, Coregonus clupeaformis, in Northern Lake Michigan as assessed by isozyme electrophoresis. Canadian Journal of Fisheries and Aquatic Science 37: 783-793.

Klumb R.A., Rudstam L. G., Mills E. L., Schneider C. P., Sawyko P. M. 2003. Importance of Lake Ontario embayments and nearshore habitats as nurseries for larval fishes with emphasis on alewife (Alosa pseudoharengus). Journal of Great Lakes Research 29(1): 181-198.

Knebelsberger T., Dunz A.R., Neumann D., Geiger M.F., Molecular diversity of Germany’s freshwater fishes and lampreys assessed by DNA barcoding. Molecular Ecology Resources 2014, 1-11.

Ko, H.L., Wang, Y.T., Chiu, T.S. et al. 2013. Evaluating the Accuracy of Morphological Identification of Larval Fishes by Applying DNA Barcoding. PLoS One, 8, 1. 1-7.

Lampert W. 1989. The adaptive significance of diel vertical migration of zooplankton. Functional Ecology 3: 21-27.

Leslie J. K. 1986. Nearshore contagion and sampling of freshwater larval fish. Journal of Plankton Research 8(6): 1137-1147.

Loftus D. H. 1980. Interviews with Lake Huron commercial fisherman. Lake Huron Fisheries Assessment Unit, Ontario Ministry of Natural Resources, Owen Sound, Ontario. Report 1-80. 634pp.

Lu G., Basley D.J., Bernatchez L. 2001. Contrasting patterns of mitochondrial DNA and microsatellite introgressive hybridization between lineages of lake whitefish (Coregonus clupeaformis); relevance for speciation. Molecular Ecology 10:965-985.

Mackie I.M., Pryde S.E., Gonzales-Sotelo C. et al. 1999. Challenges in the identification of species of canned fish. Trends in Food Science and Technology 10(1): 9-14.

Meier R., Zhang G., Ali F. 2008. The use of mean instead of smallest interspecific distance exaggerates the size of the ‘barcoding gap’ and leads to misidentification. Systematic Biology 57: 809–813.

Moritz C., Cicero C. 2004. DNA barcoding: promise and pitfalls. PLoS Biology 2(10):e354.

Moser H. G., Smith P. E. 1993. Larval fish assemblages and oceanic boundaries. Bulletin of Marine Science 53(2): 283-289.

220 Paine, M. A., McDoweel J. R., Graves J. E. 2007. Specific identification of Western Atlantic ocean scombrids using mitochondrial DNA cytochrome c oxidase subunit 1 (COI) gene region sequences. Bulletin of Marine Science 80(2): 353-367.

Pannard A., Planas D., Beisner B. E. 2015. Macrozooplankton and the persistence of the deep chlorophyll maximum in a stratified lake. Freshwater Biology. Doi:10/1111/fwb.12604.

Pfrender, M.E., Hawkins, C.P., Bagley, M. et al. 2010. Assessing Macroinvertebrate biodiversity in freshwater ecosystems: Advances and Challenges in DNA-based approaches. Quarterly Review of Biology 85: 319-340.

Quinteiro J.C., Sotelo G., Rehbein H. et al. 1998. Use of mtDNA Direct Polymerase Chain Reaction (PCR) Sequencing and PCR-Restriction Fragment Length Polymorphism Methodologies in Species Identification of Canned Tuna. Journal of Agriculture and Food Chemistry. 46(4):1662-1669.

Stocco L. B., Joyeux J.-C. 2015. Distribution of fish larvae on the Vitoria-Trindade Chain, southwestern Atlantic. Check List 11(2): 1590. 11pp.

Stott W., VanDeHey J.A., Sloss B.L. 2010. Genetic diversity of lake whitefish in lakes Michigan and Huron; sampling standardization, and research priorities. Journal of Great Lakes Research 36: 59-65.

Stott W., Ebener M. P., Mohr L., Hartman T., Johnson J., Roseman E. F. 2013. Spatial and temporal genetic diversity of lake whitefish (Coregonus clupeaformis (Mitchell)) from Lake Huron and Lake Erie. Advances in Limnology. 64: 205-222.

Swanson, S., J.F. Archibald and G. Muecke. 2015. Joint Review Panel environmental assessment report - Deep Geologic Repository for Low and Intermediate-level Radioactive Waste Project (May 6, 2015). CEAA Reference No. 17520, Canadian Environmental Assessment Agency, Ottawa, Ontario, Canada. 32pp.

Taylor W. W., Freeberg M. H. 1984. Effect of food abundance on larval lake whitefish, Coregonus clupeaformis Mitchell, growth and survival. Journal of Fish Biology 25: 733- 741.

VanDeHey J. A., Sloss B. L., Peeters P. J., Sutton T. M. 2010. Determining the efficacy of microsatellite DNA-based mixed stock analysis of Lake Michigan’s lake whitefish commercial fishery. Journal of Great Lakes Research 36: 52-58.

Young M. K., McKelvey K. S., Pilgrim K. L., Schwartz M. K. 2013. DNA barcoding at riverscape scales: assessing biodiversity among fishes of the genus Cottus (Teleostei) in northern Rocky Mountain streams. Molecular Ecology Resources 1-13.

221